Sleep
12(3):277-286, Raven Press, Ltd., New York
© 1989 Association of Professional Sleep Societies
Effects of Dream Reflection on Waking Affect:
Awareness of Feelings, Rorschach Movement,
and Facial EMG
Tore A. Nielsen, Don L. Kuiken, and Diane L. McGregor
Department of Psychology, University of Alberta, Edmonton, Alberta, Canada
Summary: The clinical and research literature suggests that waking dream reflection increases awareness of feelings. To examine this possibility, 16 male
and 16 female participants spent a single night in the sleep laboratory while
rapid eye movement (REM) sleep physiology and facial electromyogram
(EMG) were monitored. Participants in the dream imagery condition were
awakened from REM sleep and asked to reflect on their dreams. Participants
in the fantasy imagery condition were awakened from REM sleep and asked to
reflect on guided fantasies modeled on dream narratives. Participants were
asked to reflect either on the affective connotations of their imagery or on the
cognitive aspects of their imagery. Unexpectedly, self-reported awareness of
feelings was less common during either affective or cognitive reflection on
dreams than during such reflection on fantasies, especially for females. Also,
kinesthetic sensitivity, as indicated by Rorschach movement responses, was
less after dream reflection than after reflection on fantasy imagery, again especially for females. In contrast, facial EMG activity was greater during dream
reflection than during fantasy reflection. Since facial EMG during REM sleep
and during dream reflection was correlated with self-reported orienting within
dream imagery, these results are understandable as evidence for the "carryover" of REM orienting activity during dream reflection, resulting in reduced
affective/kinesthetic sensitivity and reduced awareness offeelings. Key Words:
Dreams-Emotion-REM sleep-Orienting response.
The influence of dreams on waking feelings is probably underestimated. The average
rate of recall is approximately one dream every 2-3 days (1), and recalled dreams
frequently have enduring effects on waking feelings (2). However, the processes by
which dreams influence waking feelings are largely unstudied. We have begun to explore the possibility that, because kinesthetic feedback is reduced during dream construction but restored during wakefulness, waking dream reflection enables increased
sensitivity to the affective/kinesthetic (feeling) connotations of dream imagery.
Accepted for publication August 1988.
Address correspondence and reprint requests to Dr. Don Kuiken at Department of Psychology. Biological
Sciences Building, University of Alberta, Edmonton, Alberta, Canada T6G 2E9.
277
278
T. A. NIELSEN ET AL.
There is increasing evidence that rapid eye movement (REM) sleep phasic events,
e.g., ponto-geniculo-occipital spikes, are physiologically similar to those waking reactions by which attention is redirected toward novel or affectively significant stimuli (3).
Specifically, during wakefulness, the orienting reaction involves temporary suppression of activity in kinesthetic pathways (4), inhibition of ongoing (nonorienting) movements (5), and the initiation of affective/exploratory movements, e.g., facial registration
of surprise and eye movements (6). Similarly, during REM sleep, orienting activity
involves suppression of kinesthetic feedback (7), inhibition of the postural musculature
(8), and muted affective/exploratory movements, e.g., twitches of the facial musculature and REMs (3).
The phenomenology of dreams is compatible with this physiologic picture. Suppression of kinesthetic feedback during REM sleep may be responsible for the infrequent
reports of kinesthetic sensations and affective reactions in representative dreams
(9,10). In addition, kinesthetic inhibition during REM sleep may be experienced within
dreams as weakness or inability to move (11) or as the absence of affect when, by
waking standards, affect would be expected (12). We recently found that people who
frequently experienced weakness or inability to move within dream events also reported that their dreams influenced subsequent waking feelings (13). One interpretation
of these results is that dreams constructed during particularly intense REM orienting
activity are accompanied by acute reductions in kinesthetic feedback. Acutely reduced
kinesthetic feedback is experienced as absent affect and as inability to move within the
dream. During later waking dream reflection, the affective/exploratory movements
originally occurring during dream construction may be reactivated-but now accompanied by kinesthetic feedback. The result is increased awareness of the affective/
kinesthetic (feeling) connotations of the dream.
Similarly, the orienting activity that occurs during REM dream construction is more
intense than during construction of waking imagery. Therefore, during dream reflection, the reactivation of REM affective/exploratory movements, with kinesthetic feedback, should increase awareness of the affective/kinesthetic significance of dream imagery. In contrast, during reflection on narratively comparable waking imagery, the
relative lack of affective/exploratory movements, even with kinesthetic feedback,
should provide relatively little opportunity for increases in awareness of feeling.
This hypothesis seems consistent with the continued use of dreams to encourage
feeling expression in psychotherapy (14). The few empirical studies available also bolster the notion that dream reflection increases access to feelings. In one series of
studies (15,16), intensification of waking affect followed revisualization of dreams.
Alpha blocking, eye movements, cardiac acceleration, and galvanic skin response indicated greater arousal immediately after dream revisualization than after either fantasy
construction or fantasy recall. In another investigation (17), clients waiting to enter
psychotherapy were given the opportunity to recall and discuss dreams that had been
reported during the previous night in a sleep laboratory. These clients expressed significantly more emotion and disclosed significantly more personal material to their
therapists during subsequent sessions of psychotherapy than did clients recalling and
discussing less dreamlike content.
However, the design of the preceding experiments did not clarify the processes
responsible for the greater awareness of feelings after dream reflection. In particular,
the differences between dream and fantasy reflection could be attributed to (a) contrasting structures of dream and fantasy narratives, (b) contrasting modes of attention
Sleep, Vol. 12, No.3, 1989
DREAM REFLECTION AND WAKING AFFECT
279
evoked by dreams and fantasies, or (c) differential temporal lags between the occurrence of and reflection on dream or fantasy imagery. In the present research, we
attempted a conceptual replication of these prior studies with an experimental design
which (a) matched the narrative structures of dream and fantasy imagery, (b) manipulated attentiveness to the affective connotations of the imagery during reflection, and
(c) held constant the time between construction of and reflection on the imagery.
Unexpectedly, the present study indicated that orientation-induced kinesthetic inhibition may persist during dream reflection which immediately follows awakening,
thereby reducing sensitivity to affectively relevant kinesthetic information. Rather than
the expected increase in awareness of feelings during dream reflection (as indicated by
self-report) and rather than the expected increase in kinesthetic sensitivity (as indicated
by Rorschach movement responses), dream reflection immediately after awakening
was associated with less affective involvement than reflection on comparable waking
imagery.
METHODS
Participants
Sixteen males (aged 17-31 years, mean = 23.3 years) and 16 females (aged 18-38
years, mean = 24.6 years) were selected from volunteers responding to a campus-wide
request for dream research volunteers. These volunteers were moderately high dream
recallers (at least one dream/week). All respondents were interviewed to determine
whether they had undergone recent psychologic distress or sleep difficulties. Two
volunteers were encouraged not to participate because of current distress.
Procedure
Participants arrived at the laboratory -1 h before normal bedtime. During electrode
application, participants were told that the purpose of the electrodes was to detect
electrophysiologic indicators of REM sleep; they were not told that some electrodes
were used for monitoring feeling-related activity of the facial musculature.
The first of two scheduled awakenings occurred after at least 5 min into either the
second or third REM sleep period. Dream reports were collected at this time to fulfill
what participants perceived to be a requirement of the study. The second scheduled
awakening occurred at least 10 min into either the fourth or fIfth REM sleep period, as
close to the participant's normal wake-up time as possible. At this time, half of the
participants were asked to recall a dream and half were provided instructions for
construction of fantasy imagery.
Dream versus fantasy imagery
Immediately after the second awakening, participants were given one of two imagery
tasks in which they either recalled their immediately preceding dream in detail or
constructed and recalled fantasies which were modeled on dream narratives. Care was
taken to equate the time and effort required to generate the imagery sequences in the
two conditions: Participants in the dream imagery condition were asked to recall the
dream without reporting it, recall it again, and then report it, and participants in the
fantasy condition were asked to construct the fantasy, recall it, and then report it.
Furthermore, fantasy instructions were designed to maximize the narrative similarity
of the guided fantasies and spontaneous dream imagery. Each dream reported by a
Sleep, Vol. 12, No.3, 1989
280
T. A. NIELSEN ET AL.
participant in the dream imagery condition was rewritten as an instruction set for use
with one sex-matched participant in the fantasy imagery group according to the following guidelines: (a) possessive pronouns and associative references from the original
dreams (e.g., "I am in my kitchen.") were replaced with pronouns and references that
were appropriate for the listening participant (e.g., "Imagine that you are in your
kitchen. "); (b) character and setting descriptions from the original dreams (e.g., "I was
with my friend Jane at the Cineplex.' ') were replaced with thematically similar characters and settings (e.g., "Imagine that you are with a female friend at a movie
theater. "); (c) characters and settings were always previewed briefly at the start of each
imagery task, allowing time for participants to embellish these more general elements
with personally relevant details.
Apart from the preceding constraints, report length, plot changes, unexplained scene
shifts, and other anomalies of imagery construction comprising the original dreams
were not modified in the fantasy instruction sets. Finally, free association to the dream
or fantasy was not encouraged, minimizing the possibility that associated memories
would mediate effects on waking feelings.
After reporting a dream, participants in the dream condition were asked to select the
dream segment that seemed most important and to rate their feelings during this segment using the Differential Emotions Scale (DES), a 30-item checklist of emotional
adjectives (c. E. Izard, F. E. Dougherty, B. M. Bloxom, and W. E. Kotsch, unpublished observations, 1974). After reporting their fantasies, participants in the fantasy
condition rated their feelings during the fantasy segments analogous to those selected
by their yoked participants in the dream condition.
Affective versus cognitive reflection
Next, participants were guided through one of two 15-min instructional sequences
designed to encourage intensive reflection on the selected imagery segment. Affective
reflection instructions encouraged participants to focus on and privately characterize
the quality of any bodily feelings that occurred as they reviewed their dreams or
fantasies. These instructions were adapted from those developed by Gendlin (14) for
focusing on bodily feelings related to dream imagery. The instructions consisted of a
series of eight directives, each followed by a quiet interval which could be terminated
either by the participant when he/she was "ready," or by the experimenter after a
preset time limit (between 45 and 90 s for different intervals).
Cognitive reflection instructions encouraged participants to attend to any thoughts
and ideas that occurred as they reviewed their imagery. The structure of the cognitive
reflection instructions was like that of the affective reflection instructions except that
when the affective reflection instructions drew attention to bodily feelings (e.g., "Pay
attention to how your body reacts and feels. "), cognitive reflection instructions drew
attention to mental events (e.g., "Pay attention to any thoughts or ideas that occur to
you. ").
Rorschach movement
Lerner (18) suggested that Rorschach movement responses indicate increased utilization of kinesthetic sensations in self-perception. In the present study, Rorschach
movement scores were used as an index of this type of kinesthetic sensitivity. Immediately after the reflection task, the experimenter presented participants with Rorschach cards I, II, III, IV, VII, and IX according to Exner's (19) procedures. A judge,
blind to condition, scored the protocols for the frequency of all verbs depicting moveSleep, Vol. 12, No.3, 1989
DREAM REFLECTION AND WAKING AFFECT
281
ment by humans, except that postural verbs were excluded. Proportion of agreement
with a second judge was 82%. Protocols were also scored on a more inclusive measure
of kinesthetic imagery consisting of the frequency of all verbs depicting movement by
humans, animals, or inanimate objects, again excluding postural verbs. Proportion of
agreement on the total movement measure was 75%.
Post-focusing inventory
Following Rorschach administration, participants were administered a 26-item PostFocusing Inventory which has been used to discriminate different types of reactions
during self-reflection instructions (20). The six-item Affective Insight subscale (e.g., "I
came face to face with feelings I usually ignore. ") was chosen for analysis as a measure
of the awareness of feelings during reflection.
Facial EMG
Electro-oculogram (EOG), electroencephalogram (EEG), and facial electromyogram
(EMG) were recorded throughout the experiment. Four electrode pairs (interelectrode
distance = 2 cm center to center) were dedicated to facial EMG. Initial test impedances
were as close to 1 kG as possible, and data from three participants for whom EMG
electrode impedances were >5 k!l in the morning were deleted from the final analysis.
Facial sites were selected to optimize detection of covert negative and positive affective qualities. Previous research (e.g., 21-24) has suggested (a) that negative affect
is accompanied by activity of right and left corrugator supercilli (RC and LC), i.e., the
muscle groups responsible for the "brow-knitting" response during sadness; and (b)
that positive affect is accompanied by activity of right and left zygomaticus major (RZ
and LZ), i.e., the muscle groups responsible for smiling during happiness. EMG potentials at these sites were sampled and digitized at a rate of 256 Hz/channel.
Phasic muscle events, assessed only during REM sleep, were defined as waveforms
which exceeded a minimum amplitude of 35 I.l. V for a duration ~0.25 s. Phasic EMG
scores for each participant were calculated as proportions of the REM period, i.e., as
the number of phasic events in the REM sleep period divided by the length of the period
in minutes. These measures were labeled RZ(REM), LZ(REM), RC(REM), and
LC(REM).
Tonic muscle amplitude was assessed only during wakefulness. Tonic amplitude was
defined as the average rectified amplitude during each of the eight intervals of the
reflection instructions. These measures were labeled RZ(WAKE), LZ(WAKE),
RC(W AKE), LC(WAKE). To minimize eye movement artifact, phasic muscle events
and tonic muscle amplitudes were computed only when EOG amplitudes were <30 I.l. V.
RESULTS
Post-focusing inventory
A subscale from the Post-Focusing Inventory, the Affective Insight Scale, was intended to measure the extent to which awareness of feelings occurred during imagery
reflection. Scores on this subscale were submitted to a 2 x 2 x 2 analysis of variance
(ANOV A) with sex, imagery type, and reflection type as between participants factors
(n = 32). There was a marginally significant sex X imagery type interaction indicating
that females in the dream imagery condition reported experiencing less awareness of
feelings (mean = 2.01) than females in the fantasy imagery condition (mean = 2.79) or
Sleep, Vol. 12, No.3, 1989
282
T. A. NIELSEN ET AL.
than males in either the dream (mean = 2.58) or fantasy (mean = 2.50) conditions, (F
(1,24) = 3.17, p < 0.09).
To clarify these results, we analyzed a second self-report measure of feeling awareness, specifically, the mean intensity rating of all 30 items from the DES for the selected
imagery segment. Analysis of this measure revealed that, even before administration of
the imagery reflection instructions, dream feelings (mean = 1.62) were significantly
less intense than fantasy feelings (mean = 1.95), (F(l, 24) = 5.99, p < 0.03). In sum,
analysis of the Affective Insight Scale and of the DES suggested less pronounced affect
in the dream condition than in the fantasy condition, perhaps especially for female
participants.
Rorschach movement
Rorschach movement responses were intended to measure sensitivity to affective/
kinesthetic sensations after imagery reflection. Scores for the Rorschach response
measures were submitted to 2 x 2 x 2 ANOVAs with sex, imagery type, and reflection
type as between-participants factors. For movement by humans, a significant sex X
imagery interaction revealed that females in the dream condition described fewer actions by humans (mean = 1.88) than females in the fantasy condition (mean = 4.75),
although this difference for males was small (mean = 3.75 vs. mean = 2.75), (F(1, 24)
= 4.32, p < 0.05).
A main effect was also found for the total number of movement responses (i.e.,
human, animal, and inanimate movement), (F(1, 24) = 5.59, p < 0.03). Participants in
the dream condition provided fewer references to movement (mean = 4.50) than participants in the fantasy condition (mean = 7.25). The total movement measure also
revealed a significant sex X imagery type interaction, (F(1, 24) = 7.21, p < 0.02). This
interaction was due both to infrequent reference to movement by females in the dream
condition (mean = 3.25) and to frequent reference to movement by females in the
fantasy condition (mean = 9.13), as compared with more moderate frequencies by
males in both reflection conditions (mean = 5.75 vs. 5.38). In sum, the Rorschach
results suggested less sensitivity to affectivelkinesthetic events in the dream reflection
condition than in the fantasy condition, especially among female participants.
Facial EMG
Tonic facial EMG was intended to provide an index of the extent to which feelingspecific motor activity would occur during imagery reflection. Measures of tonic facial
EMG during waking reflection, i.e., RZ(WAKE), LZ(WAKE), RC(WAKE), and
LC(WAKE), were submitted to 2 x 2 x 2 x 8 ANOV As with sex, imagery type (dream
vs. fantasy), and reflection type (affective vs. cognitive) as between participant factors
and reflection interval (1-8) as a within participant factor (n = 29).
A significant imagery type X reflection interval interaction for RZ(WAKE) revealed
that, regardless of reflection type, dream imagery resulted in greater motor activity than
did fantasy imagery during certain intervals, (F(7, 133) = 2.63, p < 0.02). As evident
from Table 1, the interaction was largely attributable to higher RZ(WAKE) amplitudes
during reflection on dream imagery than during reflection on fantasy imagery for interval two ("What is the most important feeling [idea] in this imagery?") and for
interval four ("Find a word or phrase that seems to fit this whole feeling [seems to
summarize this whole idea]. "). For LZ(WAKE), a similar but statistically insignificant
pattern was evident.
Sleep, Vol. 12, No.3, 1989
283
DREAM REFLECTION AND WAKING AFFECT
TABLE 1. Average rectified EMG amplitudes (microvolts) during successive reflection
intervals for two types of imagery and four different facial muscle sites
Interval
EMG
Imagery
RZ
Dream
Fantasy
LZ
Dream
Fantasy
RC
Dream
Fantasy
LC
Dream
Fantasy
2
3
4
5
6
7
8
4.49
3.36
16.50
3.92
4.84
3.45
10.00
4.50
6.42
3.57
4.88
3.99
4.58
3.71
4.34
4.71
5.03
4.43
11.09
4.93
4.79
4.57
9.59
5.49
5.35
4.62
5.16
5.05
4.62
4.90
4.45
5.47
5.71
4.97
10.79
5.86
5.46
5.82
6.98
6.10
6.51
5.80
6.57
7.49
6.55
6.24
6.96
6.74
5.91
4.48
8.80
5.00
6.21
4.43
6.81
4.95
6.16
4.50
6.14
5.56
6.46
4.57
6.27
5.02
RZ, right zygomaticus major; LZ, left zygomaticus major; RC, right corrugator supercilli; LC, left corrugator supercilli.
For RC(WAKE) , a pattern like that obtained for RZ(WAKE) occurred during interval 2, but a sex X imagery type X reflection interval interaction indicated that the effect
was specific to female participants, (F(7, 133) = 2.22, P < 0.04). That is, for females
during interval 2, RC(WAKE) amplitudes were greater in the dream condition (mean =
16.35) than in the fantasy condition (mean = 5.41). For males, this difference (mean =
6.03 vs. 6.19) was negligible. For LC(WAKE) , a similar but statistically insignificant
interaction (p < 0.09) was observed.
In sum, there was evidence that reflection on dream imagery led to greater activation
of facial musculature than did reflection on fantasy imagery. This effect was most
robust for RZ(WAKE) during reflection intervals in which attention was directed toward the most important personal feeling or ideas in the imagery segment.
Supplementary analyses
Results from the Affective Insight Scale and for Rorschach movement suggested that
dream reflection attenuated rather than intensified affective/kinesthetic experience.
These results seemed inconsistent with the finding that facial muscle activity (especially
RZ) was more intense during reflection on dream imagery than during reflection on
fantasy imagery. However, examination of the correlations between RZ(REM),
RZ(WAKE), and DES ratings in the selected imagery segments suggested an alternative interpretation of the psychological significance of the facial EMG measures. We
devised an orienting activity subscale by adding the DES ratings for attention and
surprise. As shown in Table 2, both (a) phasic EMG activity during REM sleep and (b)
tonic EMG activity during waking dream reflection correlated significantly with orienting activity in the dream imagery condition but not in the fantasy imagery condition.
No other DES subscales differentiated the two imagery conditions in this manner.
Similar but statistically unreliable results were found for LZ but not for RC or LC EMG
sites.
These data suggested that RZ(WAKE) was an index of orienting activity during
dream reflection, rather than reactivated affect as originally expected. In addition,
since orienting reactions suppress kinesthetic feedback, and since orienting activity, as
Sleep, Vol. 12, No.3, 1989
T. A. NIELSEN ET AL.
284
TABLE 2. Correlations between RZ EMG and the
DES-orienting response subscale for two types of imagery
Dream
Stage
a
RZ (REM) x orienting
RZ (WAKE) x orientin~
Fantasy
0.02
0.16
EMG, electromyelogram; DES, Differential Emotions Scale.
a Dream and fantasy correlations differ: p < 0.05, one-tailed.
b p < 0.01, One-tailed.
indicated by RZ(W AKE), was apparently intensified at times during dream reflection,
we suspected that RZ(WAKE) would correlate negatively with measures of (a) awareness offeelings and (b) sensitivity to affective/kinesthetic events. Indeed, RZ(WAKE)
correlated -0.57 with the Affective Insight Scale in the dream condition and +0.11 in
the fantasy condition, correlations which differ significantly from each other (p < 0.05,
one-tailed). Similarly, RZ(W AKE) correlated - 0.35 with Rorschach movement by
humans in the dream condition and + 0.37 in the fantasy condition, correlations which
again differ significantly from each other (p < 0.05, one-tailed).
DISCUSSION
Contrary to our original expectations, participants in the dream reflection condition
were less likely to report awareness of feelings and less likely to provide Rorschach
movement responses than participants in the fantasy reflection condition. Supplementary analyses suggested that the attenuation of affective/kinesthetic experience may be
a particular form of carry-over effect, i.e., persistence or reactivation of REM orienting
activity during waking. Specifically, in the dream condition only, DES ratings of orienting activity correlated positively with both RZ(REM), a measure of REM sleep
facial muscle activity, and RZ(W AKE) , a measure of waking facial muscle activity.
Therefore, the intensification of RZ(WAKE) during certain dream reflection intervals
may be understood as the carry-over of REM sleep orienting activity. That RZ(WAKE)
amplitudes were highest during dream reflection intervals calling for assessment of
imagery significance is consistent with this interpretation, since orienting activity is
frequently a response to the perceived significance of a stimulus (25). That the EMG
site most clearly involved was RZ is also compatible with a recent study (26) in which
orienting responses to a visual array included mouth movements, i.e., movements
likely to affect EMG in the zygomatic musculature.
Imagery conditions in the present study may be usefully compared with conditions in
a study of carry-over effects conducted by Piss and colleagues (27) who showed that
thematic apperception test (TAT) fantasies elicited during the post-REM period contained more emotion than TAT fantasies elicited during wakefulness. The experimental
task of Piss and colleagues is most similar to the fantasy condition of the present study.
In both studies, immediate post-REM tasks involved creating fantasies, which in the
case of Piss and colleagues were structured by TAT cards and in the present case were
structured by guided fantasy instructions. In both studies, immediate post-REM fantasies were also more affectively involving than in other conditions.
However, the present results suggest that the findings of Piss and colleagues may be
better described as the disinhibition of affect enabled by shifting attention away from
dream content per se to related images, memories, etc. Piss and colleagues found
Sleep, Vol. 12, No.3, 1989
DREAM REFLECTION AND WAKING AFFECT
285
that intensity of affect in the post-REM fantasies was significantly correlated with the
number of REMs, an index of how motorically active the prior REM period was. We
found that intensity of some feelings (happiness, disgust) in post-REM fantasies was
significantly correlated (r = 0.59 and r = - 0.45, respectively) with another index of
prior REM state motor activation, i.e., phasic zygomatic activity. Thus, there is evidence of carry-over in both studies. However, the component of REM activity which
persists or is reactivated during fantasy reflection may not be the same as that component carried over during direct reflection on the dream. The present data suggest (a)
that during direct dream reflection immediately after awakening from REM sleep, REM
orienting activity is carried over into waking, reducing awareness of feelings; and (b)
that during reflection on dream-related fantasies or memories immediately after awakening from REM sleep, dreams' affective connotations are released from inhibition and
carried over into waking.
This interpretation of the present data and that of Fiss and colleagues (28) bears
directly on the studies of dream reflection by Morishige and Reyher (15,16) and Cartwright and associates (17). These investigators reported immediate affective arousal
and continuing expression of affect as a function of dream reflection. However, their
experimental paradigms are perhaps more like the fantasy condition than the dream
condition of the present study. In the studies of Morishige and Reyher, participants
were encouraged to reveal "hot images" associated with their dreams. In the study of
Cartwright and associates participants focused on and discussed their dreams in the
morning, thereby relating their dreams to waking memories, etc. In neither case was
reflection on the dream limited to the dream per se, as it was in the dream condition of
the present study. Consequently, their results may be due to carry-over of dreams'
affective connotations during the processes by which dream mentation is related to
waking images and memories, rather than to direct reflection on dream content per se.
The preceding account partly depends on the validity of post hoc analyses of data
from the present study and require further validation. Little is known about processes
responsible for the sex difference obtained here. Future studies should attempt clarification of the dream reflection processes that release the normally inhibited affective
connotations of dreams.
Acknowledgment: This work was supported by Grant No. A-8040 from the Natural Sciences
and Engineering Research Council to D. L. Kuiken.
REFERENCES
1. Webb WB, Kersey J. Recall of dreams and the probability of stage I-REM sleep. Percept Mot Skills
1967;24:627-30.
2. Wasserman I, Ballif BL. Perceived interactions between the dream and waking divisions of consciousness. J [mag Cog Pers 1984;4:3-13.
3. Morrison AR, Reiner PB. A dissection of paradoxical sleep. In: McGinty DJ, et aI., eds. Brain mechanisms of sleep. New York: Raven Press, 1985:97-110.
4. Brunia CHM. Some questions about the motor inhibition hypothesis. In: Kimmel HD, Van OIst EH,
Orlebeke JF, eds. The orienting reflex in humans. Hillsdale, NJ: Lawrence Erlbaum, 1979:241-58.
5. Kahneman D. Attention and effort. Englewood Cliffs, NJ: Prentice-Hall, 1973.
6. Rorbaugh JW. The orienting reflex: Performance and central nervous system manifestations. In: Parasuraman R, Davies DR, eds. Varieties of attention. New York: Academic Press, 1984:323-73.
7. pompeiano O. Mechanisms responsible for spinal inhibition during desynchronized sleep: experimental
study. In: Weitzman ED, ed. Advances in sleep research, vol 3. New York: Spectrum, 1976:411-49.
8. Glenn LL. Brain stem and spinal control of lower limb motor neurons with special reference to phasic
Sleep, Vol. 12, No.3, 1989
286
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
T. A. NIELSEN ET AL.
events and startle reflexes. In: McGinty DJ, et aI., eds. Brain mechanisms of sleep. New York: Raven
Press, 1985:81~95.
Knapp PH. Sensory impressions in dreams. Psychoanal Q 1956;25:325-47.
Snyder F. The phenomenology of dreaming. In: Madow H, Snow LH, eds. The psychodynamic implications of the physiological studies on dreams. Springfield, IL: Charles C Thomas, 1970: 124-51.
Morrison AR, Dinges DF. Reports of inhibitory motor experiences in a normal young adult population.
Sleep Res 1986;15:84.
Foulkes D, Sullivan B, Kerr NH, Brown L. Appropriateness of dream feelings to dreamed situations.
Cog Emot 1988;2:29-39.
Kuiken D, Nielsen T, Chahley S. Orientation-induced inhibition and the influence of dreams on waking
activities. Sleep Res 1988;17:103.
Gendlin ET. Let your body interpret your dreams. Wilmette, IL: Chirion, 1986.
Morishige H, Reyher J. Alpha rhythm during three conditions of visual imagery and emergent uncovering
psychotherapy: the critical role of anxiety. J Abnorm PsychoiI975;84:531-8.
Reyher H, Morishige J. Electroencephalogram and rapid eye movements during free imagery and dream
recall. J Abnorm Psychol 1969;74:576--82.
Cartwright RD, Tipton LW, Wicklund J. Focusing on dreams. Arch Gen Psychiatry 1980;37:275-7.
Lerner B. Rorschach movement and dreams: a validation study using drug-induced dream deprivation.
J Abnorm Psychol 1966;71:75-86.
Exner JE. The Rorschach: a comprehensive system, vol. 1. New York: Wiley, 1974.
Kuiken D, Carey R, Nielsen T. Moments of affective insight: their phenomenology and relations to
selected individual differences. J [mag Cog Pers 1987;6:341--64.
Cacioppo JT, Petty RE, Losch ME, Kim HS. E1ectromyographic activity over facial muscle regions can
differentiate the valence and intensity of affective reactions. J Pers Soc Psycho I 1986;50:260-8.
Fridlund AJ, Izard CEo Electromyographic studies of facial expressions of emotions and patterns of
emotions. In: Cacioppo JT, Petty RE, eds. Social psychophysiology: a sourcebook. New York: Guilford
Press, 1983:243-86.
Schwartz GE, Fair PL, Salt P, Mandel MR, Klerman GL. Facial muscle patterning during affective
imagery in depressed and nondepressed subjects. Science 1976;192:489-91.
Schwartz GE, Brown SL, Ahem GL. Facial muscle patterning and subjective experience during affective
imagery: sex differences. Psychophysiology 1980;17:75-82.
Bernstein AS. The orienting response as novelty and significance detector: reply to O'Gorman. Psychophysiology 1979;16:263-73.
Karis D, Druckman D, Lissak R, Donchin E. A psychophysiological analysis of bargaining: ERPs and
facial expressions. Ann NY Acad Sci 1984;425:230-5.
Fiss H, Klein GS, Bokert E. Waking fantasies following interruption of two types of sleep. Arch Gen
Psychiatry 1966;14:543-51.
Sleep, Vol. 12, No.3, 1989