In search of an objective measure of the process of falling asleep, and motivated by an article on blinking in the wake-sleep transition , in 1953 Kleitman and Aserinsky described the presence of rapid eye movements (REMs) in deep sleep [16, 17]. They observed that REMs appear approximately 90 min after sleep onset, are associated with desynchro-nized cortical low-voltage fast EEG activity and a significant increase and instability in heart and respiratory rates, and occur at regular intervals throughout the whole night. To test the hypothesis that REM sleep may be linked with dreaming, ten probands were awakened during periods with and without REMs. Vivid dreams were reported in 74% when awakened during periods with REMs, but in only 17% when awakened in the absence of REMs.
The neurosurgeon and neurophysiologist Jouvet discovered muscle atonia of REM sleep in 1959 . The combination of REMs, activated EEG and muscle atonia suggested the term "paradoxical sleep" as a synonym of REM sleep. Jouvet also demonstrated that the generation of REM sleep depends on an intact pontine tegmentum  and that REM atonia is due to an inhibition of motor centers in the medulla oblongata. Cats with lesions around the locus pericoeruleus exhibit a variety of complex behaviors during REM sleep, including motor patterns suggesting attack, defense and exploration .
Another co-worker of Kleitman, Dement, elaborated further on the relationship between REM sleep and dreaming. He was convinced that dreams only occur during REM sleep, and suggested a direct connection between dream content, REMs and motor activities during REM sleep [21, 22]. In line with Freud's theory, he also suggested, on the basis of a REM sleep deprivation study (which later was strongly criticized), that REM sleep was essential for psychological health .
In the 1970s, following the pioneer work of Jouvet, Hobson et al. further elucidated the brainstem-related neurophysiological and neurochemical machinery of REM sleep. They described cholinergic "REM-on" neurons and monoaminergic "REM-off" neurons (producing norepineph-rine and serotonin) in the pontine tegmentum, and REM sleep as the result of their reciprocal interaction (Fig. 1) . These authors developed a new dream theory called the "activation-synthesis hypothesis," according to which dreams are the result of random/nonspecific activation provided by the brainstem, and subsequent, more or less coherent interpretation (synthesis) by the forebrain .
Fig. 1. Subcortical and cortical brain activation/deactivation during REM-sleep based on positron emission tomography (PET) and neurophysiological data (red, activated; blue, deactivated). Brain areas: PT, pontine tegmentum; A/P, amygdala and parahippocampal cortex; AC, anterior cingulum; PC, posterior cingulum; DPC, dorsolateral prefrontal cortex. Nuclei: LDT, laterodorsal tegmental nuclei; PPT, pedunculopontine nuclei
Was this article helpful?