What Is The Dement and Kleitman Study? (REM Sleep Study)
The 1957 Dement and Kleitman study examined whether rapid eye movements during sleep were linked to dreaming. Researchers monitored nine adult volunteers and woke them during REM or non-REM sleep. Participants reported dreams after about 80% of REM awakenings and 7% of non-REM awakenings. The results showed a strong association, not proof that dreaming happens only during REM sleep.
The study is often called a landmark because it gave researchers a way to compare sleep-related signals with people’s reports after waking. It also has clear limits: it involved a small group, and dream reports depend on what people remember and say. Keeping those points in view helps you understand what the study found, and what it did not.
Study overview: research question, participants, and sleep-lab method
Dement and Kleitman asked whether eye movements during sleep could help identify when people were dreaming. Their 1957 paper described a laboratory method that compared recorded sleep activity with reports collected after people were awakened. The study involved nine adult volunteers, so its findings offer evidence about those participants, not a rule for everyone.
The original paper was published by William Dement and Nathaniel Kleitman in Electroencephalography and Clinical Neurophysiology. Its full title was “The relation of eye movements during sleep to dream activity: An objective method for the study of dreaming.” The phrase “objective method” points to a key aim: measure activity during sleep, then compare it with reports made after waking.
What do REM and non-REM mean?
REM stands for “rapid eye movement.” It is a sleep period marked by bursts of eye movement. Non-REM means sleep outside those periods. The researchers used these categories to compare dream reports after awakenings. Neither term, by itself, tells us exactly what a person was experiencing or remembering.
The team recorded physiological activity during sleep, including eye movements. They then woke participants during REM or non-REM periods and asked whether they had been dreaming. If a participant reported a dream, researchers also collected a description of it.
This procedure linked two types of evidence:
- Recorded observations: sleep-related body activity, including eye movements, and whether the awakening occurred during REM or non-REM sleep.
- Self-reports: participants’ answers about dreaming and the accounts they gave after waking.
That distinction matters. The recording showed when certain physical activity occurred. The report showed what participants recalled and described. The researchers could compare the two, but they could not directly observe a dream.
Evidence and findings: REM awakenings, dream reports, and timing
The clearest result was the difference in dream reports after REM and non-REM awakenings. Participants reported dreams after about 80% of REM awakenings, compared with about 7% of non-REM awakenings. This large contrast supports a link between REM sleep and dream reports, but it does not mean that every REM awakening follows a dream or that no one dreams during non-REM sleep.
The percentages describe what participants reported after being awakened. They are not a direct count of all dreams that occurred. A person may not remember a dream, may not call an experience a dream, or may have trouble describing it after waking. So the safest wording is that dream reports were much more common after REM awakenings in this study.
What did the study test beyond dream reports?
Dement and Kleitman also explored whether people could tell how long a REM period had lasted and whether eye-movement patterns matched reported dream imagery. These were additional questions, not proof that eye movements provide a dependable transcript of a dream.
For the timing question, participants were awakened after differing lengths of REM activity and asked to estimate how long they had been dreaming. The study reported that participants could distinguish some shorter and longer intervals. That finding suggests a possible link between REM-period duration and perceived dream duration in the tested conditions. It should not be treated as a precise clock for dreams in general.
The researchers also compared the direction or pattern of eye movements with dream descriptions. Some reported imagery appeared to correspond to the eye-movement pattern. For example, a participant who described looking at something could show eye movements consistent with looking around. Such comparisons were suggestive, but they do not show that eye movements reliably reveal the exact content of a dream.
| Study question | What researchers compared | Main point |
|---|---|---|
| Are dreams reported after REM awakenings? | Dream reports after REM and non-REM awakenings | Reports followed about 80% of REM awakenings and 7% of non-REM awakenings |
| Can participants judge dream-period length? | Estimates after REM periods of differing lengths | The study reported that participants could distinguish some durations |
| Do eye movements match dream imagery? | Recorded eye-movement patterns and later descriptions | Some patterns appeared to correspond; this did not make dream content directly observable |
The figures in the first row are the study’s reported results, not universal rates. They apply to the participants and methods in the 1957 paper.
Academic evaluation: strengths, limitations, and student learning
The study’s strength was its close comparison between sleep recordings and reports collected after awakenings. Its limits include a small sample, reliance on memory and self-report, and a design that shows association rather than cause. Reading the results with both sides in mind helps prevent a common mistake: turning a strong pattern into an absolute claim.
What were the study’s main strengths?
First, the researchers recorded activity during sleep rather than relying only on people’s general memories the next morning. Second, they awakened participants in different sleep periods and asked direct questions soon afterward. This made it possible to compare dream reports with the sleep state at the time of waking.
The study also separated several questions that are sometimes blurred together. The frequency of dream reports, estimates of dream duration, and possible links between eye movements and imagery are related, but they are not the same result. Keeping them separate makes summaries more accurate.
What are the main limitations?
- Small sample: Nine adults took part. A group this small cannot represent all people, ages, sleep patterns, or dream experiences.
- Limited generalizability: The findings describe a particular sleep-lab study. They do not establish that the same rates apply to everyone.
- Self-report: Dreaming was assessed through what participants said after waking. Their accounts could be shaped by recall, interpretation, or difficulty putting an experience into words.
- Association, not causation: REM awakenings were linked to more dream reports. The study did not show that REM caused the dreams or was required for them.
- Awakening effects: Waking someone to ask questions is part of the method, but it also means the reports were collected under specific laboratory conditions.
A careful student answer might say: “Dement and Kleitman found that participants were much more likely to report dreams when awakened during REM sleep than during non-REM sleep.” That sentence matches the evidence. “They proved that all dreams happen in REM sleep” does not.
When I explain the study, I use a simple distinction: a signal and a report are related evidence, not the same thing. If a participant reports a dream after a REM awakening, the report supports a link. It does not let a researcher look inside the dream or confirm every detail independently.
Legacy and relevance: how the study shaped dream research
Dement and Kleitman’s work mattered because it showed how sleep-lab measures could be paired with reports gathered at planned times. That approach made dream research more systematic. The lasting lesson is not that one sleep stage explains every dream, but that clear methods can test focused questions while leaving room for uncertainty.
Before interpreting a study like this, check four things: who took part, what researchers measured, what participants reported, and how far the conclusion reaches. Here, the sample was nine adults; researchers recorded sleep-related activity and used awakenings; participants reported whether they had been dreaming; and the strongest conclusion is an association between REM sleep and dream reports.
This distinction is useful in introductory psychology and research methods. A result can be meaningful without being final. The study offered a practical way to investigate dreaming, while its sample and reliance on reports limited what could be concluded.
The article’s central takeaway is simple: dream reports were far more common after REM awakenings in this study, but the study did not prove that dreaming occurs only during REM sleep.
Frequently asked questions about the Dement and Kleitman study
What was the aim of the Dement and Kleitman study?
The researchers tested whether eye movements during sleep were related to dream activity. They compared sleep recordings with participants’ reports after waking during REM or non-REM periods.
When was the study published?
It was published in 1957 in Electroencephalography and Clinical Neurophysiology. The paper’s title was “The relation of eye movements during sleep to dream activity: An objective method for the study of dreaming.”
How many people took part?
Nine adult volunteers took part. That small sample is an important limit when considering whether the findings apply to people more broadly.
What did participants report after REM awakenings?
Participants reported dreams after about 80% of REM awakenings. This is a rate of reported dreams in the study, not a direct measure of every dream that occurred.
What did participants report after non-REM awakenings?
They reported dreams after about 7% of non-REM awakenings. This finding means dream reports were less common after non-REM awakenings, not that dreaming was impossible then.
Did the study prove that dreams happen only during REM sleep?
No. Participants reported dreams after some non-REM awakenings. The study found that dream reports were much more common after REM awakenings.
Did eye movements reveal exactly what people dreamed?
No. The researchers found some apparent links between eye-movement patterns and reported imagery. Those links did not show that researchers could reliably read a dream’s exact content from eye movements.
Could participants estimate how long they had been dreaming?
The study tested estimates after REM periods of differing lengths and reported that participants could distinguish some durations. This does not establish that people can accurately time every dream.
What is the main limitation of the study?
Its small sample and reliance on self-reports limit the conclusions. Dream accounts depend on what people remember and describe after waking.
Why is the study still taught?
It is a clear example of researchers pairing physiological recordings with reports from participants. It also helps students practice separating a study’s findings from claims that go beyond its evidence.
(This article was written by one of our staff writers, Nathan Sterling. Visit our Meet the Team page.)
