EEG Lempel-Ziv complexity varies with sleep stage, but does not seem to track dream experience

Abstract

In a recent electroencephalography (EEG) sleep study inspired by complexity theories of consciousness, we found that multi-channel signal diversity progressively decreased from wakefulness to slow wave sleep, but failed to find any significant difference between dreaming and non-dreaming awakenings within the same sleep stage (NREM2). However, we did find that multi-channel Lempel-Ziv complexity (LZC) measured over the posterior cortex increased with more perceptual ratings of NREM2 dream experience along a thought-perceptual axis. In this follow-up study, we re-tested our previous findings, using a slightly different approach. Partial sleep-deprivation was followed by evening sleep experiments, with repeated awakenings and immediate dream reports. Participants reported whether they had been dreaming, and were asked to rate how diverse, vivid, perceptual, and thought-like the contents of their dreams were. High density (64 channel) EEG was recorded throughout the experiment, and mean single-channel LZC was calculated for each 30 s sleep epoch. LZC progressively decreased with depth of non-REM sleep. Surprisingly, estimated marginal mean LZC was slightly higher for NREM1 than for wakefulness, but the difference did not remain significant after adjusting for multiple comparisons. We found no significant difference in LZC between dream and non-dream awakenings, nor any significant relationship between LZC and subjective ratings of dream experience, within the same sleep stage (NREM2). The failure to reproduce our own previous finding of a positive correlation between posterior LZC and more perceptual dream experiences, or to find any other correlation between brain signal complexity and subjective experience within NREM2 sleep, raises the question of whether EEG LZC is really a reliable correlate of richness of experience as such, within the same sleep stage.

Keywords: EEG; complexity; consciousness; dreaming; sleep.

FIGURE 1

Distribution of sleep stages, dream experience classifications, and subjective ratings. (A) Total number of scored 30 s sleep epochs from each sleep stage (summed over all participants and trials). (B) Number of subjective reports collected from each sleep stage (stage of last epoch before alarm) for eyes closed (EC) and sleep trials (reports were not collected from EO trials). (C) Number of subjective reports from each sleep stage classified as no experience (NE), dream experience without recall of contents (DEWR) and dream experience with recall of contents (DE). (D) Distribution of subjective ratings from NREM2 DE awakenings, showing how diverse, vivid, perceptual, and thought-like NREM2 dream contents were, on a scale from 0 to 10 (see section “2.4. Dream report questionnaire” for detailed explanation of the scale).

FIGURE 2

EEG channel selections. Whole-brain (A) and posterior (B) channel selections. Electrode fill color indicates associated cortical lobe (C) [adapted from illustration by Laurens R. Krol, distributed under a CC0 1.0 license (Krol, 2020)].

FIGURE 3

Variation in whole-brain average single-channel Lempel-Ziv complexity (LZC) with sleep stage. (A) Average single-channel LZC vs. sleep stage for the whole-brain channel selection (all channels, see Figure 2A). Each data point corresponds to one 30 s sleep epoch. Observations are randomly scattered along the x-axis to reduce overlap, and participant number is indicated by marker fill color. Overall mean values for each sleep stage are indicated by black diamond markers. (B) Whole-brain average single-channel LZC vs. sleep stage (0 = W, 1 = NREM1, 2 = NREM2, 3 = NREM3, 4 = REM), plotted separately for each study participant. Each data point corresponds to one 30 s sleep epoch. Fill color indicates sleep stage, and black diamond markers indicate participant mean values for each stage.

FIGURE 4

Posterior average single-channel Lempel-Ziv complexity (LZC) versus NREM2 dream experience classification. Average single-channel LZC vs. NREM2 dream experience class (NE, no experience; DEWR, dream experience without recall of contents; DE, dream experience), for the posterior EEG channel selection (see Figure 2B). Each data point corresponds to the last 30 s sleep epoch before an awakening from NREM2 sleep. Observations are plotted on top of corresponding boxplots. Participant number is indicated by marker fill color, and observations are displaced slightly along x-axis to avoid overlap.

FIGURE 5

Average posterior single-channel Lempel-Ziv complexity (LZC) versus subjective ratings of NREM2 dream experience. Posterior LZC vs. ratings of how diverse (A), vivid (B), and perceptual (C) NREM2 dream experience was. Each data point corresponds to the last 30 s sleep epoch before an awakening from NREM2 sleep. Participant number is indicated by marker fill color, and observations are displaced slightly along x-axis to avoid overlap. Linear trend for each participant (for which there is more than one data point) is indicated by background line segments (visual aid only).

FIGURE 6

Lempel-Ziv complexity (LZC) versus the aperiodic spectral exponent within each sleep stage. Average LZC over all channels plotted against average aperiodic spectral exponent for Wake, NREM1, NREM2, NREM3, and REM sleep. Marker color indicates participant number. Overall mean LZC and overall mean aperiodic spectral exponent (calculated across participants) for each sleep stage is indicated by a black cross.

FIGURE 7

Channel-wise differences in Lempel-Ziv complexity (LZC) in different sleep stages compared to wakefulness. Topoplot of channel-wise differences in estimated marginal mean LZC for stages NREM1 (A), NREM2 (B), NREM3 (C), and REM (D) versus Wake. For each EEG channel, we fit a linear mixed model (LMM) with sleep stage as a fixed factor (including intercept), and participant and trial (nested within participants) as random intercepts. Epochs were entered as repeated measures with first-order auto regressive residual variance-covariance (same as the model for average LZC over all channels as a function of sleep stage). Approximate EEG electrode positions are indicated by black dots, and channels for which LZC was significantly different from wake at the un-adjusted 0.05 level are marked by white circles.

FIGURE 8

Channel-wise differences in Lempel-Ziv complexity (LZC) in NREM2 awakenings without (explicit recall of) dream experience compared to awakenings with dream experience. Topoplot of channel-wise differences in estimated marginal mean LZC for awakenings without dream experience (A) and awakenings without recall of dream experience contents (B), versus awakenings with (explicit recall of) dream experience. For each EEG channel, we fit a linear mixed model (LLM) with dream experience classification as a fixed factor (including intercept) and participant as a random intercept (same as the model for average LZC over posterior channels as a function of dream experience classification). Approximate EEG electrode positions are indicated by black dots, and channels for which LZC was significantly different from dream experience (DE) at the un-adjusted 0.05 level are marked by white circles (there are none).

FIGURE 9

Channel-wise correlation between dream experience content ratings and Lempel-Ziv complexity (LZC) for NREM2 awakenings with dream experience. Channel-wise correlation between LZC and subjective ratings of how diverse (A), vivid (B), perceptual (C), and thought-like (D) NREM2 dream experiences were. For each subjective rating, and for each EEG channel, we fit a linear mixed model (LMM) with the subjective rating of dream contents as a fixed covariate (including intercept) and participant as a random intercept. Approximate EEG electrode positions are indicated by black dots, and channels for which the subjective ratings were significant covariates at the un-adjusted 0.05 level are marked by white circles (there are none).

FIGURE 10

Distribution of subjective ratings of dream experience for each sleep stage. Distribution of subjective ratings of how diverse (A), vivid (B), perceptual (C), and thought-like (D) dream experiences were for each of the five sleep stages (Wake, NREM1, NREM2, NREM3, and REM). Each data point corresponds to the report from one dream experience (DE) awakening. Observations are plotted on top of corresponding boxplots. Participant number is indicated by marker fill color, and observations are displaced slightly along x-axis to avoid overlap.