The multiple time scales of sleep dynamics as a challenge for modelling the sleeping brain

Beyond rhythm - a framework for understanding the frequency spectrum of neural activity

Cognitive and behavioral processes are often accompanied by changes within well-defined frequency bands of the local field potential (LFP i.e., the voltage induced by neuronal activity). These changes are detectable in the frequency domain using the Fourier transform and are often interpreted as neuronal oscillations. However, aside some well-known exceptions, the processes underlying such changes are difficult to track in time, making their oscillatory nature hard to verify. In addition, many non-periodic neural processes can also have spectra that emphasize specific frequencies. Thus, the notion that spectral changes reflect oscillations is likely too restrictive. In this study, we use a simple yet versatile framework to understand the frequency spectra of neural recordings. Using simulations, we derive the Fourier spectra of periodic, quasi-periodic and non-periodic neural processes having diverse waveforms, illustrating how these attributes shape their spectral signatures. We then show how neural processes sum their energy in the local field potential in simulated and real-world recording scenarios. We find that the spectral power of neural processes is essentially determined by two aspects: (1) the distribution of neural events in time and (2) the waveform of the voltage induced by single neural events. Taken together, this work guides the interpretation of the Fourier spectrum of neural recordings and indicates that power increases in specific frequency bands do not necessarily reflect periodic neural activity.

Phenotypic Interindividual Differences in the Dynamic Structure of Sleep in Healthy Young Adults

Introduction

Evaluating the dynamic structure of sleep may yield new insights into the mechanisms underlying human sleep physiology.

Methods

We analyzed data from a 12-day, 11-night, strictly controlled laboratory study with an adaptation night, 3 iterations of a baseline night followed by a recovery night after 36 h of total sleep deprivation, and a final recovery night. All sleep opportunities were 12 h in duration (22:00-10:00) and recorded with polysomnography (PSG). The PSG records were scored for the sleep stages: rapid eye movement (REM) sleep; non-REM (NREM) stage 1 sleep (S1), stage 2 sleep (S2), and slow wave sleep (SWS); and wake (W). Phenotypic interindividual differences were assessed using indices of dynamic sleep structure - specifically sleep stage transitions and sleep cycle characteristics - and intraclass correlation coefficients across nights.

Results

NREM/REM sleep cycles and sleep stage transitions exhibited substantial and stable interindividual differences that were robust across baseline and recovery nights, suggesting that mechanisms underlying the dynamic structure of sleep are phenotypic. In addition, the dynamics of sleep stage transitions were found to be associated with sleep cycle characteristics, with a significant relationship between the length of sleep cycles and the degree to which S2-to-W/S1 and S2-to-SWS transitions were in equilibrium.

Discussion

Our findings are consistent with a model for the underlying mechanisms that involves three subsystems - characterized by S2-to-W/S1, S2-to-SWS, and S2-to-REM transitions - with S2 playing a hub-like role. Furthermore, the balance between the two subsystems within NREM sleep (S2-to-W/S1 and S2-to-SWS) may serve as a basis for the dynamic regulation of sleep structure and may represent a novel target for interventions aiming to improve sleep.

Changes in cortical, cardiac, and respiratory activities in relation to spontaneous rhythmic jaw movements in ketamine-anesthetized guinea pigs

It has been reported that rhythmic jaw movements (RJMs) spontaneously occur in ketamine-anesthetized animals. The present study investigated the physiological processes that occur during the cortical, cardiac, and respiratory events which contribute to the genesis of RJMs in animals after supplemental ketamine injections. Fourteen guinea pigs were prepared to allow electroencephalographic, electrocardiographic, and electromyographic activities to be recorded from the digastric muscle, measurement of jaw movements, and nasal expiratory airflow under ketamine-xylazine anesthesia. Rhythmic jaw movements spontaneously occurred with rhythmic digastric muscle contractions, 23-29 minutes after injection of supplemental ketamine (12.5 and 25.0 mg kg^-1 , intravenously). The cycle length of RJMs did not differ significantly between the two doses of ketamine (mean±SD: 12.5 mg kg^-1 , 326.5 ± 60.0 ms; 25 mg kg^-1 , 278.5 ± 45.1 ms). Following injection of ketamine, digastric muscle activity, heart and respiratory rates, and cortical beta power significantly decreased, while cortical delta and theta power significantly increased. These changes were significantly larger in animals given 25.0 mg kg^-1 of ketamine than in those given 12.5 mg kg^-1 . With the onset of RJMs, the levels of these variables returned to pre-injection levels, regardless of the dose of ketamine administered. These results suggest that, following supplemental ketamine injections, spontaneous RJMs occur during a specific period when the pharmacological effects of ketamine wear off, and that these RJMs are characterized by stereotypical changes in cardiac, respiratory, and cortical activities.

Estimation of Time-Varying Spectral Peaks and Decomposition of EEG Spectrograms

Detection of spectral peaks and estimation of their properties, including frequency and amplitude, are fundamental to many applications of signal processing. Electroencephalography (EEG) of sleep, in particular, displays characteristic oscillations that change continuously throughout the night. Capturing these dynamics is essential to understanding the sleep process and characterizing the heterogeneity observed across individuals. Most sleep EEG analyses rely on either time-averaged spectra or bandpassed amplitude/power. Unfortunately, these approaches obscure the time-variability of peak properties, require specification of a priori criteria, and cannot distinguish power from nearby oscillations. More sophisticated approaches, using various spectral models, have been proposed to better estimate oscillatory properties, but these too have limitations. We present an improved approach to spectrogram decomposition, tracking time-varying parameterized peak functions and dynamically estimating their parameters using a modified form of the iterated extended Kalman filter (IEKF) that incorporates discrete On/Off-switching of peak combinations and a sampling step to draw the initial reference trajectory. We evaluate this approach on two types of simulated examples-one nearly within the model class and one outside. We find excellent performance, in terms of spectral fits and accuracy of estimated states, for both simulation types. We then apply the approach to real EEG data of sleep onset, obtaining quality spectral estimates with estimated peak combinations closely matching the expert-scored sleep stages. This approach offers not only the ability to estimate time-varying parameters of spectral peaks but, moving forward, the potential to estimate the governing dynamics and analyze their variability across nights, subjects, and clinical groups.

Automatic detection of microsleep episodes with feature-based machine learning

STUDY OBJECTIVES

Microsleep episodes (MSEs) are brief episodes of sleep, mostly defined to be shorter than 15 s. In the electroencephalogram (EEG), MSEs are mainly characterized by a slowing in frequency. The identification of early signs of sleepiness and sleep (e.g. MSEs) is of considerable clinical and practical relevance. Under laboratory conditions, the maintenance of wakefulness test (MWT) is often used for assessing vigilance.

METHODS

We analyzed MWT recordings of 76 patients referred to the Sleep-Wake-Epilepsy-Center. MSEs were scored by experts defined by the occurrence of theta dominance on ≥1 occipital derivation lasting 1-15 s, whereas the eyes were at least 80% closed. We calculated spectrograms using an autoregressive model of order 16 of 1 s epochs moved in 200 ms steps in order to visualize oscillatory activity and derived seven features per derivation: power in delta, theta, alpha and beta bands, ratio theta/(alpha + beta), quantified eye movements, and median frequency. Three algorithms were used for MSE classification: support vector machine (SVM), random forest (RF), and an artificial neural network (long short-term memory [LSTM] network). Data of 53 patients were used for the training of the classifiers, and 23 for testing.

RESULTS

MSEs were identified with a high performance (sensitivity, specificity, precision, accuracy, and Cohen's kappa coefficient). Training revealed that delta power and the ratio theta/(alpha + beta) were most relevant features for the RF classifier and eye movements for the LSTM network.

CONCLUSIONS

The automatic detection of MSEs was successful for our EEG-based definition of MSEs, with good performance of all algorithms applied.

Automatic Sleep Staging Employing Convolutional Neural Networks and Cortical Connectivity Images

Understanding of the neuroscientific sleep mechanisms is associated with mental/cognitive and physical well-being and pathological conditions. A prerequisite for further analysis is the identification of the sleep macroarchitecture through manual sleep staging. Several computer-based approaches have been proposed to extract time and/or frequency-domain features with accuracy ranging from 80% to 95% compared with the golden standard of manual staging. However, their acceptability by the medical community is still suboptimal. Recently, utilizing deep learning methodologies increased the research interest in computer-assisted recognition of sleep stages. Aiming to enhance the arsenal of automatic sleep staging, we propose a novel classification framework based on convolutional neural networks. These receive as input synchronizations features derived from cortical interactions within various electroencephalographic rhythms (delta, theta, alpha, and beta) for specific cortical regions which are critical for the sleep deepening. These functional connectivity metrics are then processed as multidimensional images. We also propose to augment the small portion of sleep onset (N1 stage) through the Synthetic Minority Oversampling Technique in order to deal with the great difference in its duration when compared with the remaining sleep stages. Our results (99.85%) indicate the flexibility of deep learning techniques to learn sleep-related neurophysiological patterns.

Oscillatory patterns in the electroencephalogram at sleep onset

Falling asleep is a gradually unfolding process. We investigated the role of various oscillatory activities including sleep spindles and alpha and delta oscillations at sleep onset (SO) by automatically detecting oscillatory events. We used two datasets of healthy young males, eight with four baseline recordings, and eight with a baseline and recovery sleep after 40 h of sustained wakefulness. We analyzed the 2-min interval before SO (stage 2) and the five consecutive 2-min intervals after SO. The incidence of delta/theta events reached its maximum in the first 2-min episode after SO, while the frequency of them was continuously decreasing from stage 1 onwards, continuing over SO and further into deeper sleep. Interestingly, this decrease of the frequencies of the oscillations were not affected by increased sleep pressure, in contrast to the incidence which increased. We observed an increasing number of alpha events after SO, predominantly frontally, with their prevalence varying strongly across individuals. Sleep spindles started to occur after SO, with first an increasing then a decreasing incidence and a continuous decrease in their frequency. Again, the frequency of the spindles was not altered after sleep deprivation. Oscillatory events revealed derivation dependent aspects. However, these regional aspects were not specific of the process of SO but rather reflect a general sleep related phenomenon. No individual traits of SO features (incidence and frequency of oscillations) and their dynamics were observed. Delta/theta events are important features for the analysis of SO in addition to slow waves.

Chimera states in coupled logistic maps with additional weak nonlocal topology

We demonstrate the occurrence of coexisting domains of partially coherent and incoherent patterns or simply known as chimera states in a network of globally coupled logistic maps upon addition of weak nonlocal topology. We find that the chimera states survive even after we disconnect nonlocal connections of some of the nodes in the network. Also, we show that the chimera states exist when we introduce symmetric gaps in the nonlocal coupling between predetermined nodes. We ascertain our results, for the existence of chimera states, by carrying out the recurrence quantification analysis and by computing the strength of incoherence. We extend our analysis for the case of small-world networks of coupled logistic maps and found the emergence of chimeralike states under the influence of weak nonlocal topology.

Sleep Modelling across Physiological Levels

Sleep and circadian rhythms are regulated across multiple functional, spatial and temporal levels: from genes to networks of coupled neurons and glial cells, to large scale brain dynamics and behaviour. The dynamics at each of these levels are complex and the interaction between the levels is even more so, so research have mostly focused on interactions within the levels to understand the underlying mechanisms—the so-called reductionist approach. Mathematical models were developed to test theories of sleep regulation and guide new experiments at each of these levels and have become an integral part of the field. The advantage of modelling, however, is that it allows us to simulate and test the dynamics of complex biological systems and thus provides a tool to investigate the connections between the different levels and study the system as a whole. In this paper I review key models of sleep developed at different physiological levels and discuss the potential for an integrated systems biology approach for sleep regulation across these levels. I also highlight the necessity of building mechanistic connections between models of sleep and circadian rhythms across these levels.

Sleep Stage Transition Dynamics Reveal Specific Stage 2 Vulnerability in Insomnia

Study Objectives

Objective sleep impairments in insomnia disorder (ID) are insufficiently understood. The present study evaluated whether whole-night sleep stage dynamics derived from polysomnography (PSG) differ between people with ID and matched controls and whether sleep stage dynamic features discriminate them better than conventional sleep parameters.

Methods

Eighty-eight participants aged 21-70 years, including 46 with ID and 42 age- and sex-matched controls without sleep complaints, were recruited through www.sleepregistry.nl and completed two nights of laboratory PSG. Data of 100 people with ID and 100 age- and sex-matched controls from a previously reported study were used to validate the generalizability of findings. The second night was used to obtain, in addition to conventional sleep parameters, probabilities of transitions between stages and bout duration distributions of each stage. Group differences were evaluated with nonparametric tests.

Results

People with ID showed higher empirical probabilities to transition from stage N2 to the lighter sleep stage N1 or wakefulness and a faster decaying stage N2 bout survival function. The increased transition probability from stage N2 to stage N1 discriminated people with ID better than any of their deviations in conventional sleep parameters, including less total sleep time, less sleep efficiency, more stage N1, and more wake after sleep onset. Moreover, adding this transition probability significantly improved the discriminating power of a multiple logistic regression model based on conventional sleep parameters.

Conclusions

Quantification of sleep stage dynamics revealed a particular vulnerability of stage N2 in insomnia. The feature characterizes insomnia better than-and independently of-any conventional sleep parameter.

Developmental Changes in Sleep Oscillations during Early Childhood

Although quantitative analysis of the sleep electroencephalogram (EEG) has uncovered important aspects of brain activity during sleep in adolescents and adults, similar findings from preschool-age children remain scarce. This study utilized our time-frequency method to examine sleep oscillations as characteristic features of human sleep EEG. Data were collected from a longitudinal sample of young children (n = 8; 3 males) at ages 2, 3, and 5 years. Following sleep stage scoring, we detected and characterized oscillatory events across age and examined how their features corresponded to spectral changes in the sleep EEG. Results indicated a developmental decrease in the incidence of delta and theta oscillations. Spindle oscillations, however, were almost absent at 2 years but pronounced at 5 years. All oscillatory event changes were stronger during light sleep than slow-wave sleep. Large interindividual differences in sleep oscillations and their characteristics (e.g., "ultrafast" spindle-like oscillations, theta oscillation incidence/frequency) also existed. Changes in delta and spindle oscillations across early childhood may indicate early maturation of the thalamocortical system. Our analytic approach holds promise for revealing novel types of sleep oscillatory events that are specific to periods of rapid normal development across the lifespan and during other times of aberrant changes in neurobehavioral function.

Fast and slow Ca2+-dependent hyperpolarization mechanisms connect membrane potential and sleep homeostasis

Several lines of evidence indicate that the sleep-wake state of cortical neurons is regulated not only through neuronal projections from the lower brain, but also through the cortical neurons' intrinsic ability to initiate a slow firing pattern related to the slow-wave oscillation observed in electroencephalography of the sleeping brain. Theoretical modeling and experiments with genetic and pharmacological perturbation suggest that ion channels and kinases acting downstream of calcium signaling regulate the cortical-membrane potential and sleep duration. In this review, we introduce possible Ca²⁺-dependent hyperpolarization mechanisms in cortical neurons, in which Ca²⁺ signaling associated with neuronal excitation evokes kinase cascades, and the activated kinases modify ion channels or pumps to regulate the cortical sleep/wake firing mode.

Chimeralike states in two distinct groups of identical populations of coupled Stuart-Landau oscillators

We show the existence of chimeralike states in two distinct groups of identical populations of globally coupled Stuart-Landau oscillators. The existence of chimeralike states occurs only for a small range of frequency difference between the two populations, and these states disappear for an increase of mismatch between the frequencies. Here the chimeralike states are characterized by the synchronized oscillations in one population and desynchronized oscillations in another population. We also find that such states observed in two distinct groups of identical populations of nonlocally coupled oscillators are different from the above case in which coexisting domains of synchronized and desynchronized oscillations are observed in one population and the second population exhibits synchronized oscillations for spatially prepared initial conditions. Perturbation from such spatially prepared initial condition leads to the existence of imperfectly synchronized states. An imperfectly synchronized state represents the existence of solitary oscillators which escape from the synchronized group in population I and synchronized oscillations in population II. Also the existence of chimera state is independent of the increase of frequency mismatch between the populations. We also find the coexistence of different dynamical states with respect to different initial conditions, which causes multistability in the globally coupled system. In the case of nonlocal coupling, the system does not show multistability except in the cluster state region.

Phase-flip chimera induced by environmental nonlocal coupling

We report the emergence of a collective dynamical state, namely, the phase-flip chimera, from an ensemble of identical nonlinear oscillators that are coupled indirectly via the dynamical variables from a common environment, which in turn are nonlocally coupled. The phase-flip chimera is characterized by the coexistence of two adjacent out-of-phase synchronized coherent domains interspersed by an incoherent domain, in which the nearby oscillators are in out-of-phase synchronized states. Attractors of the coherent domains are either from the same or from different basins of attractions, depending on whether they are periodic or chaotic. The conventional chimera precedes the phase-flip chimera in general. Further, the phase-flip chimera emerges after the completely synchronized evolution of the ensemble, in contrast to conventional chimeras, which emerge as an intermediate state between completely incoherent and coherent states. We have also characterized the observed dynamical transitions using the strength of incoherence, probability distribution of the correlation coefficient, and framework of the master stability function.

Different kinds of chimera death states in nonlocally coupled oscillators

We investigate the significance of nonisochronicity parameter in a network of nonlocally coupled Stuart-Landau oscillators with symmetry breaking form. We observe that the presence of nonisochronicity parameter leads to structural changes in the chimera death region while varying the strength of the interaction. This gives rise to the existence of different types of chimera death states such as multichimera death state, type I periodic chimera death (PCD) state, and type II periodic chimera death state. We also find that the number of periodic domains in both types of PCD states decreases exponentially with an increase of coupling range and obeys a power law under nonlocal coupling. Additionally, we also analyze the structural changes of chimera death states by reducing the system of dynamical equations to a phase model through the phase reduction. We also briefly study the role of nonisochronicity parameter on chimera states, where the existence of a multichimera state with respect to the coupling range is pointed out. Moreover, we also analyze the robustness of the chimera death state to perturbations in the natural frequencies of the oscillators.

Spatial coexistence of synchronized oscillation and death: A chimeralike state

We report an interesting spatiotemporal state, namely the chimeralike incongruous coexistence of synchronized oscillation and stable steady state (CSOD) in a network of nonlocally coupled oscillators. Unlike the chimera and chimera death state, in the CSOD state identical oscillators are self-organized into two coexisting spatially separated domains: In one domain neighboring oscillators show synchronized oscillation and in another domain the neighboring oscillators randomly populate either a synchronized oscillating state or a stable steady state (we call it a death state). We consider a realistic ecological network and show that the interplay of nonlocality and coupling strength results in two routes to the CSOD state: One is from a coexisting mixed state of amplitude chimera and death, and another one is from a globally synchronized state. We provide a qualitative explanation of the origin of this state. We further explore the importance of this study in ecology that gives insight into the relationship between spatial synchrony and global extinction of species. We believe this study will improve our understanding of chimera and chimeralike states.

Engineering a thalamo-cortico-thalamic circuit on SpiNNaker: a preliminary study toward modeling sleep and wakefulness

We present a preliminary study of a thalamo-cortico-thalamic (TCT) implementation on SpiNNaker (Spiking Neural Network architecture), a brain inspired hardware platform designed to incorporate the inherent biological properties of parallelism, fault tolerance and energy efficiency. These attributes make SpiNNaker an ideal platform for simulating biologically plausible computational models. Our focus in this work is to design a TCT framework that can be simulated on SpiNNaker to mimic dynamical behavior similar to Electroencephalogram (EEG) time and power-spectra signatures in sleep-wake transition. The scale of the model is minimized for simplicity in this proof-of-concept study; thus the total number of spiking neurons is ≈1000 and represents a "mini-column" of the thalamocortical tissue. All data on model structure, synaptic layout and parameters is inspired from previous studies and abstracted at a level that is appropriate to the aims of the current study as well as computationally suitable for model simulation on a small 4-chip SpiNNaker system. The initial results from selective deletion of synaptic connectivity parameters in the model show similarity with EEG power spectra characteristics of sleep and wakefulness. These observations provide a positive perspective and a basis for future implementation of a very large scale biologically plausible model of thalamo-cortico-thalamic interactivity-the essential brain circuit that regulates the biological sleep-wake cycle and associated EEG rhythms.

Chimera death: symmetry breaking in dynamical networks

For a network of generic oscillators with nonlocal topology and symmetry-breaking coupling we establish novel partially coherent inhomogeneous spatial patterns, which combine the features of chimera states (coexisting incongruous coherent and incoherent domains) and oscillation death (oscillation suppression), which we call "chimera death". We show that due to the interplay of nonlocality and breaking of rotational symmetry by the coupling, two distinct scenarios from oscillatory behavior to a stationary state regime are possible: a transition from an amplitude chimera to chimera death via in-phase synchronized oscillations and a direct abrupt transition for larger coupling strength.

Sleep, neuroengineering and dynamics

Modeling of consciousness-related phenomena and neuroengineering are fields that are rapidly growing together. We review recent approaches and developments and point out some promising directions of future research: Understanding the dynamics of consciousness states and associated oscillations, pathological oscillations as well as their treatment by stimulation, neuroprosthetics and brain-computer-interface approaches, and stimulation approaches that probe, influence and strengthen memory consolidation. In all these fields, computational models connect theory, neurophysiology and neuroengineering research and pave a way towards medical applications.

Sleep and the functional connectome

Sleep and the functional connectome are research areas with considerable overlap. Neuroimaging studies of sleep based on EEG-PET and EEG-fMRI are revealing the brain networks that support sleep, as well as networks that may support the roles and processes attributed to sleep. For example, phenomena such as arousal and consciousness are substantially modulated during sleep, and one would expect this modulation to be reflected in altered network activity. In addition, recent work suggests that sleep also has a number of adaptive functions that support waking activity. Thus the study of sleep may elucidate the circuits and processes that support waking function and complement information obtained from fMRI during waking conditions. In this review, we will discuss examples of this for memory, arousal, and consciousness after providing a brief background on sleep and on studying it with fMRI.

The sleeping brain as a complex system

'Complexity science' is a rapidly developing research direction with applications in a multitude of fields that study complex systems consisting of a number of nonlinear elements with interesting dynamics and mutual interactions. This Theme Issue 'The complexity of sleep' aims at fostering the application of complexity science to sleep research, because the brain in its different sleep stages adopts different global states that express distinct activity patterns in large and complex networks of neural circuits. This introduction discusses the contributions collected in the present Theme Issue. We highlight the potential and challenges of a complex systems approach to develop an understanding of the brain in general and the sleeping brain in particular. Basically, we focus on two topics: the complex networks approach to understand the changes in the functional connectivity of the brain during sleep, and the complex dynamics of sleep, including sleep regulation. We hope that this Theme Issue will stimulate and intensify the interdisciplinary communication to advance our understanding of the complex dynamics of the brain that underlies sleep and consciousness.