M. Steriade: Neuronal Substrates of Sleep and Epilepsy 
E. F. Pace‐Schott, M. Solms, M. Blagrove and S. Harnad (eds): Sleep and Dreaming: Scientific Advances and Reconsiderations

Cambridge University Press, 2003. $120.00, 522pp, ISBN 0-521-81707-2 Cambridge University Press, 2003. $33.00, 360 pp, ISBN 0-521-00869-7 Recently, Cambridge University Press published two interesting books on sleep. Both are state-of-the-art, but they are very different in concept. The first, by Mircea Steriade, is a monograph. The second is a reprint of five target articles with peer-review comments that appeared earlier (2000) in the discussion journal Behavioral and Brain Sciences plus an update on relevant papers published between 2000 and 2002. Other differences are that Steriade's book is a paradise mainly for neurophysiologists, while Sleep and Dreaming contains theories on the relation between REM sleep and dreaming, the function of dreaming and the function of REM sleep, clearly referring more to the psychological literature. For more than 40 years Steriade has been carrying out neurophysiological experiments focused on the system that is responsible for wakefulness, sleep and REM sleep. Particularly the work performed in the last 25 years in Quebec, which was on the cutting edge of science. Steriade has been and still is the leading scientist in the field of neurophysiological studies on low-frequency oscillations such as delta sleep and sleep spindles, the activity in various networks such as the thalamo-cortical circuitry, pathology in the form of sleep related types of epilepsy and the effects of neuromodulatory systems on sleep and wake. Steriade describes his recordings from identified neurons in vivo in cats, often in chronic experiments in naturally awake or sleeping cats or in epilepsy models. Many important findings are described: he and his teams of excellent neurophysiologists discovered a new slow (<1 Hz) rhythm generated intracortically in the EEG of cats, that was later also found to be present in the human sleep EEG. Steriade was also a pioneer in the discovery of the thalamic origin (reticular thalamic nucleus, nRT) of sleep spindles, which produce hyperpolarizations leading to spike-bursts in thalamic relay cells that transfer oscillations to the cortex. Another of Steriade's remarkable points of view is that generalized and bilaterally synchronous seizures are not suddenly generalized and bilaterally synchronous. Instead they have a cortical focus and may remain local or progressively build up and recruit other parts of the brain quickly through cortico-cortico or cortico-thalamic pathways. The book is extremely interesting for neurophysiologists interested in epilepsy and sleep; for non-neurophysiologists the book gets even more exciting when Steriade discusses the function of slow wave sleep or REM sleep. Steriade's point of view is shaped by his neurophysiological vision. He dismisses Horne's view that stage two of slow wave sleep is mainly optional or a luxury. He is intrigued by the results from neurophysiological studies that demonstrate that sleep spindles and delta waves reflect long lasting inhibitory processes in cortical and thalamic neurons, during which thalamic relay nuclei prevent the transfer of information from the sensory systems to the cortex. The brain is deaf and blind so to speak, and the scene is set for a closed, sleepy brain. Steriade cannot believe that we spend so much of our lives in a useless state of unconsciousness, as assumed by Horne. Alternatively, the loss of consciousness serves as a period of recuperation for synapses. He is intrigued by the fact that, although there is no incoming signal from the thalamus, cortical GABA-ergic interneurons show unexpected high levels of neuronal activity during slow wave sleep. Based on evidence mainly from Buzsaki's group, Steriade concludes that sleep patterns in the limbic system are important for the preservation or consolidation of synaptic modifications induced by experience. Put into more popular terms, Steriade proposes that what is experienced during the day may be reactivated during slow-wave sleep in the hippocampus and also in the neocortex. The evidence for the memory consolidation hypothesis of slow wave sleep is however, still quite weak. In the classical sleep literature the pros and cons of the consolidation hypothesis for REM sleep are well known and it is often concluded that methodological problems are the main cause for the controversies. The consolidation hypothesis for slow wave sleep is even more debatable. Therefore it appears that Steriade's idea about the function of non-REM sleep is insufficiently supported by solid experimental evidence, although his ideas about the abundance of inhibitory processes during sleep and reduced possibilities of thalamo-cortical cells to transfer information are extremely relevant and solid. The good side is that his hypotheses on the function of sleep are testable. Steriade's book gives a comprehensive and unifying picture of his fantastic work, witnessing the tremendous progress in the field of neurophysiology of sleep. His book provides a great overview of the progress that has been made in the neurophysiology of sleep and various types of epilepsy. The progress in the field of the relation between sleep and dreams is terrifyingly small, as witnessed from the often-acrimonious academic debates in Sleep and Dreaming (see Allan Hobson's first remark in answer to his critics). This book contains five keynote articles. The first three are concerned with the relationships between dreaming and brain neurophysiology and neurochemistry in the framework of the relationship between REM sleep and dreaming. In all five articles the debates start with the discovery of REM sleep by Azerinsky and Kleitman in 1953. The non-REM sleep EEG, characterized by large amplitude slow waves, K-complexes and sleep spindles is periodically interrupted with a waking EEG. In addition, the discovery that dreaming preferably could be reported after awakenings from REM sleep was very exciting for those interested in the secret message that might be hidden in the dream content. The coupling of REM sleep and dreaming suggests that dreams are the function of REM sleep. However, it was discovered as early as the 1960s that dream reports also could be obtained from non-REM sleep. From then onwards, the progress appears to have stopped, the discussion about the function of REM sleep being dominated by methodological problems. Studies of dream content were hampered by the fact that dream recall upon awakening is far from reliable. A new dimension to the debate was only recently obtained from neurobiological studies applying brain-imaging techniques. While EEG studies show close similarities in the EEG of awake and REM sleep subjects, PET studies show that the dorsolateral parts of the prefrontal cortex brain are deactivated during REM sleep, in contrast to wakefulness. It might be that the somewhat bizarre dream content and the deficits in self-reflective awareness and orientation are due to the deactivation of part of the frontal cortex. This new viewpoint suggests that REM sleep constitutes a state of generalized brain activity with the specific exclusion of the prefrontal function that normally participates in high order analysis, in integration of sensory and other neural information and in behavioral inhibition. Dreaming appears to be a consequence of various forms of cerebral activation during sleep. Activation is different in different phases of sleep, so is the content of the mental processes that are the consequence of these activations. The above does not imply that we know the function of REM sleep or the function of dreams. However, the release of the handcuffs which held the association between REM sleep and dreams intact may open new pathways to investigate the function of either of them. In the fourth keynote article, the widely held belief that REM sleep has a memory consolidation function is convincingly challenged, while in the last keynote it is argued that dreams are adaptive because they increase the probability of reproductive success in threatening environments. Revonsuo proposes and defends his Threat-Simulation Theory of Dreaming. This theory is very interesting but the quality of the data that are used lacks the solidity of what can be achieved in neurobiology. In summary, it would be better if the progress in the research domains as described in Sleep and Dreaming were to come from neurobiologically oriented research. However, the various current perspectives on the function of REM sleep and dreaming are thoroughly discussed. The book is therefore seriously recommended for students of dream theory.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.