Psi man mascot

Curious Phenomena of Sleep

A variety of strange and interesting events occur during sleep. Many seem to involve partial awakening.

The brain is full of specialized subsystems or modules that can act semi-independently. Sometimes, as with sleepwalking or sleeptalking, skilled behavior can be produced while a person remains deeply asleep.

Other unique events can occur during the transition from sleep to awakening (notably sleep paralysis and hypno­pompic hallucinations). These may well be cases of partial awakening during which some parts of the brain are activated while others remain in a dream-like state.

How might curious phenomena of sleep be explained?

First we will review sleep phenomena that are odd or peculiar but not really abnormal. They occur in most people at least once or twice in their lives, sometimes more often, but are not associated with brain abnormalities or disease.

Sleepwalking and Sleeptalking

Automatic activity during sleep is called sleepwalking or somnam­bulism. Typically sleepwalking involves simple, goal-directed activity of the type one could do automatically or absent-mindedly.

The activity is carried out without normal awareness, and the sleep­walker usually has no memory of the activity later. Sleepwalking is more likely to occur under the following conditions:

1. during childhood

2. to males

3. if a person is under stress

4. during non-REM sleep (stages 3 and 4)

5. after a person drinks alcohol

What are typical characteristics of sleepwalking? When is sleepwalking likely to occur?

Sleepwalking is associated with non-REM sleep because there is no muscle inhibition as there is during REM sleep. The sleeper is able to move.

However, a sleepwalker is not neces­sarily acting out a dream. Jacobson and Kales (1967) found dreams of sleep­walkers bore no resemblance to activities they carried out while sleep­walking. That is a strange finding. It implies a radical split between two trains of activity in the nervous system.

A sleepwalker does not walk with arms outstretched or eyes closed. The facial expression may be more blank than usual. The sleepwalker is sufficiently aware of the environment to avoid obstacles, respond to verbal questions or commands, and to perform complex activities.

Many students have heard it is dangerous to awaken a sleep­walker. That is not necessarily true. It is true, however, that sleepwalking takes place during deep phases of non-REM sleep when a sleeper is difficult to awaken.

A person awakened during sleepwalking may be disoriented and groggy for up to twenty minutes. Like any other person awakened from a deep sleep, he or she may resent being awakened or be confused by sudden demands. Experts say it is better simply to lead a sleepwalker back to bed.

What is supposedly the best way to deal with a sleepwalker?

Sleeptalking is less oriented to events in the environment than sleepwalking. If you attempt to converse with a sleeptalker, the sleeptalker will usually respond with nonsense.

However, one student discovered a simple way to get a coherent response from a sleeptalking person. Call the person's name before you make your statement.

The dreamer may respond to you directly or may incorporate your voice into the dream, responding to you as if you were somebody else. Ask the wrong questions, however, and you may be playing with fire. One student reports:

I had a good observation of a sleep talker one morning. My wife, who is a long sleeper, had a tendency to talk in her sleep the first year of our marriage. Although she had done it several times, there is one exper­ience that sticks in my mind.

During a trip to Disney World, I awoke one morning to find my wife mumbling. Knowing that she talks in her sleep, I began to ask her simple questions. I received no response until I said her name first, and then asked her the same question.

I mixed up the questions, sometimes stating her name first, sometimes not. I found that she answered questions only when I stated her name. Of all the questions, my favorite answer was to the question, "What did you like best about Disney World?" The answer was, "Dumbo."

But something strange happened a short time later. I asked her if she loved me, but she would not answer the question. So I spent about two minutes telling her who I was and then asked the question again. This time I got a yes answer.

I took a break for a few minutes. I followed the break with this question: "Have you loved anyone else but Steve (my name)?" and the answer was, "No, you'll tell Steve."

This shocked me, because she would not answer questions with more than one word before. I tried to convince her that I would not tell, but she never answered the question. So I ended the questioning there.

This experience is hard to explain, but if you could ever experience it yourself, you would find it quite entertaining. (However, my wife tells me even today that I made the whole thing up.) [Author's files]

What procedure did a student suggest, to get a coherent answer from a sleeptalking person?

Unlike sleepwalking, sleeptalking is only randomly related to sleep stages. Perhaps this is because sleeptalking uses muscles located above the neck. These muscles are not inhibited during REM sleep.

The Hypnagogic State

The strange state of mind you pass through when first falling asleep is called the hypnagogic state. First studied by Baillarger in 1846 and named by Maury in 1848, hypnagogic phenomena have been experienced by about 70% of the adult population.

Common hallucinations while falling asleep are faces and geometric patterns. Such hallucinations are not a sign of mental abnormality, although they can be startling if they appear unexpectedly.

What is the hypnagogic state? What hallucinations are common during this state?

Psychologist Thore Nielsen of Hospital du Sacre-Coeur in Montreal was interested in the hypnagogic state, so he trained himself to fall asleep at his computer. That way he could wake up when he had a hypnagogic hallucination and immediately record its content.

Nielsen recorded 240 hallucinations this way. Most of the images involved movement. The most common recurring image was "falling or stepping out into space" (Adler, 1993).

This could be a measurement effect: an effect on the data caused by the nature of the measurement. Going to sleep in front of his computer, Nielsen was in constant danger of losing muscle tone and falling to the floor, which might account for repeated images of falling into space.

What causes hypnagogic imagery? It might be like an early-occurring dream fragment. The one time story-like dreams are likely to occur, besides during REM sleep, is immediately after sleep onset.

This assumption is also consistent with reports that hypnagogic images are dreamlike. A subject in a 1930s experiment on hypnagogic imagery said, "These things [hypnagogic images] are practically dreams, but I am awake enough to catch them."

The Hypnopompic State

Just as the period of transition into sleep has a name (the hypnagogic state) so does the transition from sleep to wakefulness. It is called the hypnopompic state, a label suggested by Myers in 1904.

The most characteristic feature of the hypnopompic state is disorient­ation. This can occur if a person is awakened out of deep non-REM sleep.

Sometimes a person becomes outwardly-oriented long enough to perform an action (like answer a telephone or turn off an alarm). The half-awake person may never wake up completely and may not remember doing anything later.

People can do intelligent things in a hypnopompic state. Doctors have been known to answer the phone at home, give intelligent, detailed instructions, only to be puzzled the next day when somebody thanks them for advice they do not remember dispensing.

What is the hypnopompic state? What is its characteristic feature? What have doctors been known to do?

Like sleepwalking and sleeptalking, hypnopompic disorientation seems to occur more readily after alcohol consumption. A student wrote:

One afternoon I went to Happy Hour. After drinking a couple of beers I went back to my apartment and fell asleep on my bed. The next thing I knew, the phone was ringing.

I must have been in a deep sleep because it took me a few seconds before I realized that the phone really was ringing and I wasn't just dreaming it. My first reaction was to jump out of bed real quick.

Then I experienced the weirdest feeling. I was standing there looking directly at the phone and listening to it ring, yet I stood there panicking, trying to figure out what I had to do to hear the person on the other end of the phone. It didn't even occur to me that all I had to do was pick up the phone and answer it.

After the phone had rung several times I finally thought to pick it up and answer it. By this time I was sweating from all the panicking, but I did answer the call. This whole little episode seems strange and scary to me, and I can still remember it vividly to this day. [Author's files]

Psychologist Roger Shepard of Stanford University was inspired by images which occurred for a few seconds while he was awakening. Shepard (1990) reported the events of one morning in 1970:

With eyes still closed on that morning, I suddenly saw before me an immense, luminously shimmer­ing, golden array of diamond-shaped panels separated by burnished beveled strips... (p.35)

Shepard described diamond-shaped patterns in a lattice or connected pattern. Shepard was startled to find similar descriptions in the writings of Sir John Herschel, who experimented with chloroform (an anesthetic) in the 19th Century.

Herschel reported vivid "filigree-like" images of "rhombic lattice works." Similar patterns are common in the hallucinations of the hypnagogic state (see the reference to "geometric patterns" above).

What hypnopompic experience did Shepard report and why was he startled by a 19th Century description?

Geometric patterns can even be produced in the visual field by pressing the eyeball from outside. They must reflect a property of the human visual system. Perhaps, like 3-D computer programs, neurons of the visual system code visual scenes into polygons.


Adler, T. (1993, September). Scientists have clearer view of body's descent into sleep. APA Monitor, 24, 21.

Jacobson, E., & Kales, A. (1967). Somnambulism: All night EEG and related studies. In S.S. Kety, E.V. Evarts, & H.I. Williams (Eds.). Sleep and altered states of consciousness. Baltimore, Md.: Williams & Wilkins. (pp. 424-455)

Shepard, Roger N. (1990) Mind Sights. New York: W. H. Freeman and Co.

Write to Dr. Dewey at

Don't see what you need? Psych Web has over 1,000 pages, so it may be elsewhere on the site. Do a site-specific Google search using the box below.