Reversed and increased functional connectivity in non-REM sleep suggests an altered rather than reduced state of consciousness relative to wake

Abstract

Sleep resting state network (RSN) functional connectivity (FC) is poorly understood, particularly for rapid eye movement (REM), and in non-sleep deprived subjects. REM and non-REM (NREM) sleep involve competing drives; towards hypersynchronous cortical oscillations in NREM; and towards wake-like desynchronized oscillations in REM. This study employed simultaneous electroencephalography-functional magnetic resonance imaging (EEG-fMRI) to explore whether sleep RSN FC reflects these opposing drives. As hypothesized, this was confirmed for the majority of functional connections modulated by sleep. Further, changes were directional: e.g., positive wake correlations trended towards negative correlations in NREM and back towards positive correlations in REM. Moreover, the majority did not merely reduce magnitude, but actually either reversed and strengthened in the opposite direction, or increased in magnitude during NREM. This finding supports the notion that NREM is best expressed as having altered, rather than reduced FC. Further, as many of these functional connections comprised "higher-order" RSNs (which have been previously linked to cognition and consciousness), such as the default mode network, this finding is suggestive of possibly concomitant alterations to cognition and consciousness.

Figure 1

Significant polynomial fits to functional connectivity (FC) data across wakefulness and sleep stages. Plus symbols indicate group-average FC values for a given edge, for a given stage. Units are Fisher r-to-z-transformed full-correlation values, taking into account autocorrelation. FC edges that change in the direction opposite to wakefulness polarity during NREM and return towards wakefulness FC during REM are indicated with red lines. Figures generated using MATLAB (R2019a; mathworks.com). FC edges best described by: (A) convex quadratic fits (N = 18; 13 red). (B) Concave quadratic fits (N = 18; 14 red). (C) Cubic fits (N = 1). (D) Linear fits (N = 5). Also see Supplemental Figs. S1 and S2, for FC matrices indicating the direct statistical comparison of edge FC distributions between specific stages. W wakefulness, REM rapid eye movement, NREM2 non-REM stage 2, SWS slow wave sleep.

Figure 2

Representative cartoon of the angular distances between vectors representing resting state network (RSN) functional connectivity (FC) in different sleep–wake stages. Vectors exist in multidimensional space, with the number of dimensions dependent on the number of FC edges that are in a given category (e.g., each vector exists in 91-dimensional space for the category “ALL edges”). However, only two dimensions are represented here, for illustrative purposes. Indicated angles are the degree-equivalent of the angular distances between the mean vectors for each stage (indicated as colored arrows), with the mean wakefulness vector always used as the reference point. Angles between sleep stages are not indicated, however, the statistical significance of these differences is indicated by asterisks (single/double/triple asterisks indicate p < 0.05/0.01/0.001, respectively). Note that angles between any pair of vectors actually exist in separate dimensional planes and are only represented in the same plane for illustrative purposes. Also note that angle equivalents between sleep stages are not equal to differences in the individual angles relative to wake, as such angles exist in multidimensional space and must be calculated separately. Colored triangles indicate the spread of vectors for each stage, again for illustrative purposes, as they are actually spread across multidimensional space. (A) ALL edges (N = 91 dimensions). (B) ALL HIGHER-ORDER resting state network edges (N = 70). (C) Default mode network (DMN) & Fronto-Parietal network (F-P) edges (N = 46). REM rapid eye movement, NREM2 non-REM stage 2, SWS slow wave sleep.

Figure 3

Summary of functional connectivity (FC) edge changes across wakefulness and Non-rapid eye movement (NREM). (A) FC changes between wakefulness and NREM stage 2 (NREM2). (B) FC changes between wakefulness and slow wave sleep (SWS). Circles indicate FC edges that increase (white), reduce (grey), or reverse (black) the polarity of FC during NREM, relative to wakefulness. FC edges that reduce wakefulness FC during one NREM stage, but reverse wakefulness FC during the other NREM stage are also indicated (grey-black circles). Circles overlaid on top of FC matrices for NREM2 (A) and SWS (B). Nodes reordered according to hierarchical clustering (hierarchy visualized above matrix). FC matrix colors represent Fisher r-to-z transformed correlations between nodes (taking into account autocorrelation and standard error), with 1-group t-test performed on all participants with data available for a given sleep stage. Select graphical elements of this figure (FC matrices, brain images, cluster hierarchies) were generated using the FSLNets network modeling toolbox (v0.6.3; fsl.fmrib.ox.ac.uk/fsl/fslwiki/FSLNets). SM somato-motor, sV striate visual, A auditory, DAN dorsal attention network, esV extra-striate visual, vsV ventral stream visual, DMN default mode network, ECN executive control network, l/rF-P left/right fronto-parietal.

Figure 4

Screenshots of residual gradient- and BCG artifact-corrected EEG tracings for each sleep stage used in the analyses. (A) Non-REM stage 2 (B) Slow wave sleep (C) Rapid eye movement (REM) sleep.

Figure 5

Cartoon of possible polynomial fits for functional connectivity (FC) data across wakefulness and sleep. (A) Null hypothesis (H₀); first-order polynomial, horizontal line fit. (B) Alternative hypothesis 1 (H₁); first-order polynomial, non-horizontal line fit. (C) Alternative hypothesis 2 (H₂); second-order polynomial, quadratic line fit. (D) Alternative hypothesis 3 (H₃); third-order polynomial, cubic line fit. REM rapid eye movement, NREM2 non-REM stage 2, SWS slow wave sleep.