Customer Service | Help | FAQ | PEP-Easy | Report a Data Error | About
Tip: To refine your search with the author’s first initial…

PEP-Web Tip of the Day

If you get a large number of results after searching for an article by a specific author, you can refine your search by adding the author’s first initial. For example, try writing “Freud, S.” in the Author box of the Search Tool.

For the complete list of tips, see PEP-Web Tips on the PEP-Web support page.

Fisher, C. Byrne, J. Edwards, A. Kahn, E. (1970). A Psychophysiological Study of Nightmares. J. Amer. Psychoanal. Assn., 18:747-782.

(1970). Journal of the American Psychoanalytic Association, 18:747-782

A Psychophysiological Study of Nightmares

Charles Fisher, M.D., Joseph Byrne, B.S., Adele Edwards, B.A. and Edwin Kahn, Ph.D.


1. Modern physiological methods of sleep research (REM-EEG) have shown that awakenings with anxiety occur during all stages of sleep. The most severe type of nightmare (pavor nocturnus or night terror) of children and adults arises out of the deepest stage of nondreaming sleep, Stage IV; the more severe posttraumatic nightmares are evidently of this type also. Although initiated out of Stage IV sleep, this type of severe nightmare is in some sense a "disorder of arousal" (Broughton). It is characterized by (a) a sudden, cataclysmic breakthrough of uncontrolled anxiety; (b) a sudden loud scream of blood-curdling intensity or cry for help; (c) the passing into an arousal reaction—subject is dissociated, confused, unresponsive, hallucinating; (d) a waking EEG pattern, but subject is not fully alert; (e) the subject frequently being propelled out of bed and moving through the house as though in flight; (f) intense autonomic discharge, heart rate doubling or nearly tripling in 15 to 30 seconds, and respiratory rate and amplitude increasing; (g) difficulty in retrieving content and rapid onset of amnesia for the attack.

2. Stage IV nightmare appears to arise out of a physiological void because there is no evidence of autonomic activation just prior to its violent onset, but rather the opposite, respiration, for example, slowing and the intensity of the nightmare being proportional to the duration of the Stage IV preceding it. This suggests that the deeper the sleep, the more intense the nightmare arising from it. These nightmares take place in the first few hours of sleep when most Stage IV is present, about two thirds of them during the first NREM period.

3. The Stage IV nightmare is rather rare, an incidence of only 3 to 4 percent of large numbers of schoolchildren having been reported, and the incidence among adults is even lower.

4. We do not believe the Stage IV nightmare arises out of a psychological void, though there is a marked tendency toward amnesia for content and the experience of the attack. There are two kinds of retrievable content: (a) ongoing during Stage IV, occurring simultaneously with or just prior to the scream, generally involving a single vivid scene like falling or being crushed;

(b) content that is elaborated subsequent to the scream during the confused hallucinatory state. Against the notion of a preonset psychological void is the fact that the content retrieved can be shown to be coherent, psychodynamically organized, related to the subject's pre-existing traumata and conflicts, and to the REM dreams of the same night.

5. The etiology of the Stage IV nightmare is obscure, the possible roles of severity of trauma, intensity of psychopathology, and organic factors all being in need of further investigation. The conditions for its onset appear to involve increasing ego regression brought about by progressive deepening of Stage IV sleep in the early hours of the night, an accompanying loss of reality testing, weakening of defense, appearance of primary process, etc. A marked resomatization of the anxiety response occurs, with regression to the traumatic situation, feelings of helplessness, uncontrolled anxiety and panic. There is a remarkable correlation between the severity of the nightmare and the amount and quality of the Stage IV sleep preceding it. It is suggested that deeper sleep and progressive ego regression during the minutes preceding the sudden onset of the nightmare are conducive to a massive eruption of repressed anxiety, associated with previous traumatic experiences of recent origin or regressively reactivated traumatic fixations of very early childhood.

6. We were not able to confirm Broughton's assertion that Stage IV nightmare subjects show a relative tachycardia during slow wave sleep. The fact that heart rate is either normal or decreased prior to onset speaks against the notion that the nightmare terror develops in response to the subject's awareness that his heart is beating fast.

7. It is possible to precipitate severe Stage IV nightmares in subjects prone to them simply by sounding a buzzer, the severity of the ensuing nightmare depending upon the amount of Stage IV just preceding the stimulus. This finding suggests that not all such nightmares are triggered by ongoing endogenous psychic content, but some may be triggered by an external stimulus which sets off the arousal response.

8. Most frightening arousals from sleep, subjectively experienced as nightmares, are REM anxiety dreams of varying degrees

of intensity. In the laboratory, such dreams are mild to moderate in intensity, rarely, if ever, having the overwhelming panicky quality of the Stage IV nightmare, and rarely accompanied by cries or vocalization. In 12 of 20 REM anxiety dreams, cardiorespiratory rates and the number of eye movements did not show fluctuations greater than normal. In the remaining eight, however, terminal anxiety preceding awakening could be shown to be associated with increases in autonomic activity and eye movements.

9. The REM dream thus appears to have a mechanism for tempering and modulating anxiety, for desomatizing the physiological response to it, that is, abolishing or diminishing the physiological concomitants. The mechanism is a relative one and operative in varying degrees in different anxiety dreams and in different subjects. The hypothesis of a desomatizing mechanism is supported by the observation that there are REM nightmares with marked anxiety not associated with concomitant autonomic discharge, by the fact that the absolute changes in heart and respiratory rates during REM sleep and REM nightmares are not very great and by the peaceful, nonanxious nature of REM sleep on nights when severe Stage IV nightmares occur, even though the REM dreams and such Stage IV nightmares deal with the same conflicts. The desomatization mechanism helps prevent disruption if anxiety arouses the sleeper, diminishes the intensity of secondary anxiety, aids in the task of mastering anxiety, and helps the dream to preserve REM sleep.

10. The Stage IV nightmare deals with uncontrolled anxiety; the REM dream deals with controlled anxiety. The former is a sudden, instantaneous, cataclysmic event associated psychically with a single scene which takes place simultaneously with it. REM anxiety causing awakening emerges out of a complex, prolonged dream that has been in progress, on the average, for 20 minutes. A striking finding is that subjects with severe Stage IV nightmares and many spontaneous NREM awakenings have quiet and peaceful REM sleep even though their dreams may deal with the same conflict material as the nightmares.

11. Much more than the REM dream the Stage IV nightmare resembles an acute, but reversible schizophrenic episode of a paranoid type. The coexistence or alternation of the latter with

psychosis and the possible prognostic significance of childhood Stage IV nightmare for later psychosis were discussed.

12. We conclude that the Stage IV nightmare does not serve to master anxiety, but rather represents a massive failure of the ego to control it. We agree with Freud that the posttraumatic nightmare is some sort of exception to the wish-fulfilling theory of dreams, but do not believe that it operates beyond the pleasure principle, under the domination of the repetition compulsion, nor that it supports the theory of the death instinct. We suggest that the Stage IV nightmare is not a dream at all in the ordinary sense, but a relatively rare symptom, a pathological formation of NREM sleep. The REM anxiety dream, on the other hand, is a normal phenomenon, present from infancy throughout life.

13. The problem of whether a particular childhood or adult posttraumatic state will eventuate in Stage IV or REM nightmares or both needs further investigation. The very low incidence of Stage IV nightmares at all periods of life points to the importance of the role of the relatively desomatized, controlled REM anxiety dream in the mastery of traumatic experience.

[This is a summary or excerpt from the full text of the book or article. The full text of the document is available to subscribers.]

Copyright © 2020, Psychoanalytic Electronic Publishing, ISSN 2472-6982 Customer Service | Help | FAQ | Download PEP Bibliography | Report a Data Error | About

WARNING! This text is printed for personal use. It is copyright to the journal in which it originally appeared. It is illegal to redistribute it in any form.