Most scientists of that era thought the brain merely idled in sleep. The discovery of REMs and their link to dreaming, however, showed the sleeping brain is active. This finding, described in a 1.5-page report published in the September 4, 1953, Science sparked the modern field of sleep medicine and led to the recognition and treatment of 84 discrete disorders of sleep and the sleep/wake cycle.
The Chicago researchers, Eugene Aserinsky, a graduate student in physiology, and Nathaniel Kleitman, Ph.D., chair of physiology, roused sleepers after seeing that the heart beat faster and breathing quickened—signs of emotional arousal—when REMs occurred. But they were more interested in the physiological concomitants of REMs than in their psychiatric implications.
“Aserinsky and Kleitman sneered at Freud. I thought Freud was God,” William Dement, M.D., Ph.D., recalled in a recent interview. Dement, in 1952 a second-year medical student planning a career in psychiatry, helped Aserinsky record subjects and went on to write or co-write more than 500 papers on sleep.
Aserinsky and Kleitman reported observing “rapid, jerky, and binocularly symmetrical eye movements” via electrooculograms in 20 sleeping adults. By today’s standards, their number of subjects was small: they interrogated 10 subjects in 27 REM episodes, hearing 20 dream reports. Of 23 interrogations of the same subjects while eyes were still, 19 elicited no dream recall.
Aserinsky and Kleitman thought dreaming occurred in a light stage of sleep. It was not until 1959 that Michel Jouvet, M.D., of Claude Bernard University in Lyon, France, recognized that REM sleep is distinct from both non-REM (NREM) sleep and waking; that is, it is a third state of existence. Because REM sleep involves brain activity much like that of wakefulness, but near paralysis of the body, Jouvet called it “paradoxical sleep.” NREM sleep, a time of reduced brain activation, is termed “quiet sleep.” NREM sleep does not involve motor inhibition.
While much has been learned about REM sleep since the initial research, Jouvet said in a talk at a session marking the 50th anniversary of the REM discovery at the annual meeting of the Associated Professional Sleep Societies (APSS) last summer, “the function of dreams remains an unsolved mystery.”
Aserinsky received his Ph.D. in 1953 and left Chicago, miffed by having to include Kleitman as a co-author on the paper reporting the doctoral research for which Kleitman had served as adviser. Although he and Kleitman published a 1955 paper expanding their original report, Aserinsky eschewed sleep studies for roughly the next decade. He later studied ocular activity and respiration in sleep at Jefferson Medical College in Philadelphia, where he eventually became a professor of physiology. He published 20 papers and a few book chapters. He died in 1998 at age 77 after his car veered off the road and hit a tree; Kleitman died in 1999 at age 104.
Dement meanwhile immersed himself in sleep research. In the summer of 1953, he studied 17 people with chronic schizophrenia and 18 healthy medical students at Manteno (Ill.) State Hospital. Knowing that Freud viewed dreams as a mental “safety valve,” Dement hypothesized that people with schizophrenia would not have REMs in sleep. To his surprise, the sleeping patients and the medical students experienced REMs equally often. About half the subjects with schizophrenia, he reported in 1955, recounted dreams of isolated, inanimate objects apparently hanging in space. None of the medical students had such dreams.
By 1955, when he graduated from medical school, Dement had divided sleep into non-REM (NREM) stages 1 to 4, or lightest to deepest sleep, plus REM, categories still used today. In 1957 he and Kleitman defined the normal human sleep cycle, showing REM periods range from 3 minutes early in sleep to 50 minutes at the end of sleep and recur roughly every 90 minutes. It was clear that everyone dreams several times every night, not just in response to a stressful situation, as had been hypothesized prior to the REM discovery.
Dement received his Ph.D. in 1957 and went to Mt. Sinai School of Medicine in New York City to work with Charles Fisher, M.D., a psychoanalyst and dream researcher. One of Dement’s studies there showed people deprived of “dreaming,” that is, sleep in which REMs occurred, preferentially recovered it when next permitted to sleep. Dement today is the Lowell W. and Josephine Q. Berry Professor of Psychiatry and Behavioral Sciences at Stanford University School of Medicine in Palo Alto, Calif., and director of its Center of Excellence in Sleep Disorders.
In 1959 Howard Roffwarg, M.D., a resident in psychiatry at Mt. Sinai, heard Dement lecture and started to work with him. Roffwarg and colleagues published a lengthy review article in Science in 1966 that summarized key developments in sleep research up to that time. They reported finding that newborns spend half of their sleep time in REM sleep and hypothesized that this state plays a vital role in development, perhaps by periodically stimulating the central nervous system. In a memorable phrase, they suggested that dreams “appear to be born in the brainstem but clothed in the cortex.”
In other early work, Roffwarg explored whether eye movements and muscle activity in REM sleep were correlated with dream content. In one instance, Roffwarg recalled in a recent interview, “We picked up rhythmic left-wrist and right-ankle movement. The sleeper, who was left-handed, reported he was dreaming about playing his guitar.
“This is a powerful story,” Roffwarg added. “It’s not just one system that’s involved.”
REM sleep behavior disorder offers an experiment in nature, noted Roffwarg, now a professor of psychiatry and director of sleep medicine at the University of Mississippi Medical Center in Jackson. Sleepers with this disorder, who usually have a neurological illness or injury, lack normal muscle atonia in REM sleep. Their eye movements and behavior, such as thrashing or hitting, often show a relationship to dreams. They might report, for example, “I was fighting a knife-wielding robber.”
Speculation that dreaming also occurs outside of REM periods arose soon after the discovery of REMs. While visual, storylike reports may emerge in all sleep stages, they come more often from REM sleep than from NREM sleep.
There’s a difference in formed visual imagery across the states, Allan Hobson, M.D., a professor of psychiatry at Harvard Medical School, agreed. “Visual imagery reaches a high zenith in REM sleep,” he said in a session on dreaming at this year’s APSS meeting. “This is reciprocal to thought. As internal perception increases, thought declines.”
Dream research today is torn between two paradigms, said Hobson: the old focus on dreams as interpretable, á la Sigmund Freud, and the new focus on dreaming as a state of consciousness that helps reveal how the brain is organized.
Dement sees neurocognitive research as complementary to psychological exploration, not a replacement for it. In REM sleep, he asserted, the brain creates a world all by itself. “You can see, hear, smell; you perceive the world around you to be real.
“Although dreams mirror concerns from the day, they are never exactly like the day,” he said. “But that they are meaningful is obvious.”
Just before the APSS meeting, Dement had some trouble with his computer and worried about being able to finish his talk. “For three nights in a row,” he recalled, “I dreamed about this. In one dream, I was desperately trying to find the manuscript for my talk. In another, the audience sank out of view.” Once things came together, he said, he no longer had such dreams.
Roffwarg finds scrutiny of dreams useful in psychotherapy, even in brief treatment. “I tell patients early on to write down a dream they remember and bring it in to discuss,” he said. “I ask for their associations and what’s going on in their life and mind. Occasionally, we hit gold.” ▪