Most people have heard of REM sleep, and many already know it is the stage where dreams occur, but what is the real relationship between eye movements and dreaming? Are the eye movements random, or do their movements have some relationship to the dream we are having?

It is possible to see at times the eyes of a sleeping person moving. These periods of prolonged rapid eye movements (REM) were thought by Dement and Kleitman to have some connection with dreaming.

It seems that all humans dream but dreams are easily forgotten. Freud argued that the function of dreaming was to preserve sleep by unconsciously fulfilling wishes which would otherwise upset and therefore disturb the sleeper. Freud therefore argued that dreams are the “royal road to the unconscious”, meaning that dreams allow therapists to have an insight into their clients’ unconscious thoughts.

More recently, psychologists have focused on cognitive and physiological explanations for dreaming. For example a cognitive approach might explain how dreaming is a way of dealing with our problems such as those relating to work and personal life. Whereas a physiological approach might explain dreaming as the result of random firing of neurones which create an image which we then put meaning to.

During a typical night a sleeper passes through different levels of sleep in a cyclic fashion between 5 and 7 times. Level 1 and 2 are light sleep characterised by irregular EEG patterns. Level 3 and 4 are deeper levels and are characterised by regular wave patterns. Stage 4 is called slow wave sleep or deep sleep. After stage 4 the sleeper goes back up the “sleep staircase” to stage 2 and there is a period of REM sleep lasting for about 15 to 20 minutes. These sleep states alternate during the night starting with a rapid descent into deep sleep, followed by progressively increased episodes of lighter sleep and REM sleep.

Note that the shaded areas are REM sleep

Research Methods


The relation of eye movements during sleep to dream activity – Dement, W. and Kleitman, N. (1957)


An obvious weakness of the study is its lack of ecological validity. The situation in which the participants had to sleep was unusual and could have affected their sleep patterns. Also the nature of the method of waking participants may have affected their ability to recall their dream.

A further problem with the study was the sample size. The sample size was small and only included 2 females so we could argue that the results were biased towards the dream pattern of men rather than women. Subsequent studies have found that there are large differences between individuals in the reports of dreaming during REM.

Subsequent studies have not supported Dement and Kleitman’s findings that there is a relationship between eye movements and what the person is dreaming about.

However the method was very tightly controlled. For example the researchers were able to control the location, sleeping time and the participants’ use of stimulants.

Write a PEE point evaluating the Dement and Kleitman study (either positively or negatively) for the following points:

  • experimental validity (control of variables)
  • ecological validity (realistic task)
  • external validity (generalisation)
  • ethics
  • reliability (replication, objectivity)
  • free will/determinism debate
  • sample

Alternative Study

Dement & Kleitman conducted a lab experiment. Can you think of a way to turn it into a field experiment? What would be the advantages of this?

Big Issue/s: Free Will vs Determinism Debate

Are we in control of our own actions? We think we are, but are we really? Our idea of free will could actually just be an illusion.

Human beings are physical beings, made of chemicals which obey the laws of physics like any other. Chemicals don’t have free will, so why should a big bag of chemicals like us? Sometimes it seems like determinism, the idea that we are not in control of our actions and that all of our behaviours are ‘determined’ by other forces, is the only possible conclusion.

Test Yourself 

Below are Paper 1 questions that have been asked on Dement & Kleitman (pre-2018). Learn the answers, then test yourself.


[q]From the study by Dement and Kleitman on sleep and dreaming:

(a) Identify two features of REM sleep. [2]

(b) Give one difference between REM sleep and NREM sleep. [2] [a](a)

  1. eyes move rapidly (EOG);
  2. body relatively inactive (EMG);
  3. increased EEG activity;
  4. dream more likely.


  1. REM dream more likely, NREM dream less likely;
  2. REM eyes move rapidly, NREM eyes still;
  3. REM body inactive, NREM body active;
  4. REM EEG active, NREM relatively inactive.
[q]The study by Dement and Kleitman (sleep and dreaming) used a self report method.

(a) Outline how the self report method was used in this study. [2]

(b) Describe one problem with self report data in this study. [2] [a](a) Woke participants and asked them to recall various aspects (spoken into a recording device near the bed), use of door bell, guess if 5 or 15 minutes [2]

(b) Not accurate, subjective. Ps may respond to demand characteristics / may give socially  desirable responses; may just make up a dream! [2] [q]The study by Dement and Kleitman looked at several different relationships between eye movements and dreaming. Describe two of the relationships that were investigated. [4] [a]

  • whether eye movements were related to REM dreaming
  • whether length of eye movement phase was related to length of dream
  • whether specific eye movements (vertical/horizontal) were related to dream content
  • whether dreaming only occurs in REM sleep and not in nREM sleep.
[q]From the study by Dement and Kleitman (sleep and dreaming):

(a) Describe two features of the sample. [2]

(b) Explain one disadvantage of this sample. [2] [a]

  • 9 adults
  • 7 males, 2 females
  • volunteers
[q]The study by Dement and Kleitman looked at different aspects of sleep and dreaming.

(a) Describe one of the aims of the study. [2]

(b) What were the results in relation to this aim? [2] [a]Any one from:

  • to see if REM sleep is associated with dreaming;
  • to see if dream length could be accurately estimated; (and narration);
  • to see if eye movement matches dream content.
[q]From the study by Dement and Kleitman (sleep and dreaming):

(a) Describe how eye movements were recorded. [2]

(b) Describe how brain waves were recorded. [2] [a]2 or more electrodes attached near the eyes to record corneo-retinal (electrical) potential (as the eyes moved)

[q]From the study by Dement and Kleitman (sleep and dreaming), outline two ways in which observations of the participants were made. [4] [a]Observations of eye movements:

  • 2 or more electrodes attached near the eyes to record corneoretinal (electrical) potential (as the eyes moved)

Observations of brain waves:

  • 2 or 3 electrodes attached to scalp
  • EEG run continuously (throughout sleep period) paper speed of 3 or 6 mm/sec
[q]The study by Dement and Kleitman (sleep and dreaming) collected data about dream duration and used an electroencephalograph (EEG).

(a) Describe the data for dream-duration estimates after 5 minutes and 15 minutes of REM. [2]

(b) Explain one advantage of using an EEG in this part of the study. [2] [a](a) Screen Shot 2015-05-05 at 10.25.42 AM

(b) able to measure REM sleep duration accurately, ensuring that comparisons to dream duration estimates were valid

[q]From the study by Dement and Kleitman on sleep and dreaming:

(a) Outline four controls in this study. [2]

(b) Give two reasons why it is important for studies such as Dement and Kleitman to use controls. [2] [a](a)

  • all eat normally but no alcohol or caffeine (= 1 control);
  • all have electrodes attached to head;
  • all woken by doorbell next to bed;
  • all used recording device next to bed;
  • arrive just before normal bedtime.

(b)• to maintain consistency between participants so raise reliability

  • to ensure that they are measuring the intended variable(s) not extraneous ones, so raising validity
[q]In the study by Dement and Kleitman (sleep and dreaming) an electroencephalograph (EEG) was used to detect eye movements and brain waves.

(a) Describe the EEG patterns associated with rapid eye movement (REM) sleep. [2]

(b) What advantage did being able to detect this association reliably give Dement and Kleitman? [2] [a](EEG for brainwaves)

  • low voltage (amplitude), fast (high frequency)

(EEG as EOG for eye movements)

  • REMS: 0.1–0.2; irregular;
  • bursts 1–2 to 50–100+;
  • vertical/horizontal eye movements;
  • vertical always at a minimum unless dreaming about up and down things;
  • little or much movement;
[q]In the study by Dement and Kleitman (sleep and dreaming), participants were awoken by a loud doorbell.

(a) Describe what the participants were then expected to do. [2]

(b) Explain why it was important that the doorbell was loud. [2] [a](a) state dreaming or not; relate dream content; into recorder; say 5 or 15 minutes;

(b) “To ensure immediate awakening in all levels of sleep”

So that they were no more likely to recall dreams (if they occurred) in one stage than another  (if they were more likely to forget them if woken slowly)

[q]Dement and Kleitman (sleep and dreaming) looked at the relationship between dream content and eye movements.

(a) Describe the dream content of one participant. [2]

(b) What did Dement and Kleitman conclude about the relationship between dream content and eye movements? [2] [a](a) P1: standing at bottom of cliff operating hoist and looking at climbers
P2: climbing ladders, looking up and down
P3: throwing basketballs at net
P4: two people throwing tomatoes at each other
P5: driving a car then speeding car from left


  • 4 main patterns: mainly vertical, mainly horizontal, vertical and horizontal, little/no movement
  • dream content linked to eye movements in sleep
[q](a) (a) Describe what Dement and Kleitman discovered about the occurrence of REM periods during the night. [2]

(b) Dement and Kleitman collected data from nights during which the sleepers had been woken. To what extent did they believe these findings were therefore generalisible? [2] [a](a)

  • Never during initial onset of sleep
  • Fairly regular intervals throughout the night
  • Frequency constant for each individual
  • Frequency varied between individuals (every 70–104 minutes, average 92 minutes)
  • Eyes not constantly in motion during REM periods
  • Length varied (3–50 minutes, mean 20 minutes)
  • REM periods longer later in the night

(b)It was generalisible;

  • because the results were similar to studies of uninterrupted sleep.
  • because it is a physiological process and same in everyone
[q]Use Dement and Kleitman (sleep and dreaming) to discuss the benefits of gathering quantitative data. [10] [a]
  • data such as frequency and voltage of brain waves very accurate therefore reliable [and validdata such as eye movements can be recorded in relation to brain activity to give accurate comparisons
  • accurate timing of REM and length of narrative suggest valid relationships
  • recording of simple 1/0 quantitative data (such as recall/no recall or right/wrong) is unambiguous so objective and low error rate so reliable
  • can use in statistical procedures, e.g. correlations between minutes in REM and number of words in narrative
  • but use of scientific equipment to improve reliability/validity may put participants off (e.g. hard to sleep attached to electrodes).
[q]In the study by Dement and Kleitman (sleep and dreaming) they say that dreaming can be measured objectively and that this has useful applications.

(a) Use an example to describe what is meant by ‘an objective measure’. [2]

(b) Suggest two useful applications of the objective measurement of dreaming, either ones which Dement and Kleitman suggested or any other useful application. [2] [a](a) A way to score a variable that is not affected by/is independent of personal viewpoint e.g. EEG/EOG to identify when participant is dreaming

(b)To study the effect of:

  • environmental changes
  • psychological stress
  • drugs

or any other appropriate suggestion

[q]From the study by Dement and Kleitman (sleep and dreaming):

(a) Explain why Dement and Kleitman believed that there would be a relationship between dreaming and rapid eye movements before conducting their study. [2]

(b) What two types of evidence helped Dement and Kleitman to conclude that there was a relationship between dreaming and rapid eye movements? [2] [a](a)“Such a relationship was reported by Aserinsky and Kleitman (1) who observed periods of rapid, conjugate eye movements during sleep and found a high incidence of dream recall in Ss awakened during these periods and a low incidence when awakened at other times. The occurrence of these characteristic eye movement and their relation to dreaming were confirmed in both normal Ss and schizophrenics (4),…”


  • dream content (qualitative)
  • EEGs (quantitative)
  • REMs (EOG) (quantitative)


[q]Use Dement and Kleitman (sleep and dreaming) to discuss the physiological approach. [10] [a]Physiological investigation, e.g. sleep, stages of sleep (REM, nREM), brain activity (low voltage, fast EEG when dreaming, high voltage, slow EEG, or EEG with frequent spindles with low-voltage background when not dreaming). Little dreaming early in night, more later, even if woken and resumed sleep suggesting CNS activity has a normal course which is followed regardless of momentary wakings. Relationship between eye movements and dream content (i.e. brain activity).

Typifies physiological approach to research as uses lab experiment, collects quantitative data [although also some qualitative, e.g. descriptions of dreams, which is unusual], strong controls (e.g. no caffeine, alcohol, reduction of demand characteristics by not informing of sleep stage), scientific equipment (e.g. EEGs for brain waves and eye movements) [although so few participants is unusual for physiological approach].

Go Further

Teaching Resources