Introduction in 1953, Aserinsky and Kleitman Discovered Human Rapid Eye Movement (rem) Sleep and Documented That Dream Reports Were Obtained Most Frequently When Subjects Were

IN 1953, ASERINSKY AND KLEITMAN DISCOVERED HUMAN RAPID EYE MOVEMENT (REM) SLEEP AND DOCUMENTED THAT DREAM REPORTS WERE OBTAINED MOST FREQUENTLY WHEN SUBJECTS WERE AWAKENED FROM REM SLEEP.1 Thereafter, many scientists conducted studies on dream and REM sleep and found a robust association between electrophysiologic phenomena and subjective experiences during REM sleep.2–7 This well-documented association has led to the conclusion that dream experiences are psychological manifestations generated by the neural system controlling REM sleep,8,9 yielding many innovative findings on the mind-body relationship. In contrast, researchers have also recorded dream reports from subjects upon awakening from nonREM (NREM) sleep; though the association has been shown to be weaker in comparison with that of dream reports upon awakening from REM sleep.10–15 Many human studies, together with some animal studies on REM sleep, have proposed a neural system responsible for human dream experiences. In the early pioneering studies on human REM sleep, researchers focused on the relationship between eye movements and visual experiences during REM sleep and postulated that REM and concomitant activation of the visual system of the brain account for human dream experiences.2,4 Recent neurobiologic findings obtained from animal studies have led to the current understanding that phasic signals arising from the pons and impinging upon the cortex during REM sleep might give rise to dream experiences.8,9 Some studies have focused on dream experiences during NREM sleep; however, no documented findings have yet afforded an understanding of the mechanisms of dreaming during NREM sleep. Rather, researchers have generally made the assumption that dream reports upon awakening from NREM sleep may be a consequence of recalling dream experiences from the preceding REM sleep rather than indicate the existence of NREM-specific dream experiences.7,16-17 However, only a few studies have aimed to determine whether dream reports after NREM sleep are derived from residuals of memory from the preceding REM sleep or actually arise from another type of dream experience during NREM sleep. This may be due to the methodologic limitations of the conventional intermittent awakening method, in which subjects under all-night polysomnography are awakened several times upon reaching the target sleep stage and asked about their dream experiences. However, investigation of NREM dreaming may require a method in which subjects enter a sleep period only consisting of NREM sleep separated by a sufficient period of wakefulness to exclude the influence of the preceding REM sleep. These prerequisites, however, have not been satisfied in most previous studies except for a study that examined dream experiences during a short and discrete sleep period.18 In the study reported here, we used a repeated-nap trial, in which 20-minute sleep periods separated by 40 minutes of enforced wakefulness were repeated for 78 hours to allow the