Dream recall and visual memory
Michael Schredl, Saskia Frauscher and Acram Shendi
Central Institute of Mental Health, Mannheim
The present study investigated the relations for 50 subjects between
frequency of dream recall, length of dream report, and visual memory. Whereas
the results confirmed the previously found relationship between frequency
of dream recall and visual memory, influence of visual memory on dream
report length was not found.
Since dreams mainly consist of visual images, several authors investigated
the relation of dream recall frequency to visual memory. Cory, Orniston,
Simmel, and Dainhoff (1975) reported that those of high recall performed
better on short-term and long-term memory tasks with collages (20 usual
dream themes) as stimuli. Although Lloyd (1976) and Butler and Watson (1985)
replicated this finding with different methods, two studies (Cohen, 1971;
Belicki, Hunt, & Kelly, 1978) did not yield a relationship between
dream recall freqency and visual memory. A 5-min. sequence of a film as
stimuli for testing visual memory was used by Waterman (1991). He obtained
a positive correlation (r=.24, p<.05, n=80) between dream recall frequency
and the visual memory score for his sample (45 to 75 years of age). Also,
he found a correlation of .66 between dream report length and visual memory
(p<.001, n=80). The present study was designed to investigate the relations
between visual memory, dream recall frequency, and dream report length
in greater detail.
The 50 participating subjects kept a dream diary over one week. After that
three memory tasks were carried out. First, a part of a city map with a
definite route was shown for 60 seconds. Immidiatly after the presentation
the subjects were asked to reproduce the route in the same now empty city
map. The test score is the sum of correct reproduced segments of the route
(Range from 0 to 31). Second, 20 pictures of objects were presented simultaneously
for 60 s. The subjects were asked shortly afterward to recall as many as
possible. The number of correct cited objects was used for further analysis.
These two tasks are subtests of the memory test LGT-3 (Bäumler, 1974).
Finally, a 5-min. sequence of the film 'Citizen Kane' of Orson Welles was
shown without sound. In the following 15-min. period the subjects wrote
down the details they could recall. To measure the performance of one subject,
72 objects and details were extracted from the film sequence. All reports
were compared with this list, the number of reproduced details was used
as individual test score.
The sample included 50 persons whose mean age was 23.2 yr. ñ
3.6. There were 35 women and 15 men, who, with three exceptions, were psychology
students. All but one reported at least one dream during the week.
Results and Discussion
The descriptive statistics of the measured variables are shown in table
1. Frequency of dream recall (diary) correlated significantly with the
subtest city map (r=.36, p<.005, one-tailed), almost significantly with
the film test (r=.22, p<.06, one-tailed) and not with the subtest objects
(r=-.07, ns). The average diary report length did not correlate substantially
with one of the three measures of visual memory (r=-.03 (city map), r=-.05
(objects) and r=.00 (film). Present results supported the findings that
visual memory is related to frequency of dream recall. In particular, Bäumler
(1974) stated that the city map subtest is measuring purely the visual
component of memory whereas the objects subtest is mixed up with verbal
components (writing down the names of the recalled objects). The small
correlation between freqeuncy of dream recall and memory for visual films
(Waterman, 1991) was confirmed. The clear superiority for the pure visual
memory in explaining the variance of dream recall frequency fits in with
other findings that especially visual creative skills are related to frequency
of dream recall and not other creative skills, e. g. play music, doing
handicraft/needlework (Schredl, 1995a). Since the measures of visual memory
showed very low intercorrelations (r<.20), it may be worthwhile to study
the construct visual memory in more detail with a variety of stimuli and
The effect of visual memory on report length of diary dreams could
not be replicated. Maybe the factors of age, general intelligence and visual-spatial
intelligence play an important role as mediators in the relationship between
visual memory and report length for the sample of Waterman (1991). All
three variables were significantly correlated with report length in his
study. Were they partialled out, they might reduce the correlation. The
findings of the present study suggest that the visual tasks examined here
are not at all that promising. Several investigations stress other factors
in explaining individual differences in dream report length, e.g divergent
thinking (Austin, 1971), creativity (Wood, Sebba, & Domino, 1989-90),
verbal intelligence (Schredl, 1994), sleep duration (Taub, 1972), report
style and arousal level (Rosenlicht, Maloney, & Feinberg, 1994) and
trait anxiety (Gentil & Lader, 1978). Nevertheless it seem useful to
include measures of visual memory in further research projects on this
topic, maybe visual memory is a mediator variable in the relationship between
various measures, e. g. creativity and dream report length. The findings
of an investigation of the dream life in the elderly (Schredl, 1995b) indicated
that especially verbal short term memory but not visual memory explains
differences in report length of diary dreams.
Austin, M. D. (1971) Dream recall and the bias of intellectual ability.
Nature, 231, 59-60.
Bäumler, G. (1974) Lern- und GedĄchtnistest (LGT-3). Göttingen:
Belicki, K., Hunt, H., & Kelly, P. (1978) The function of dream
and dreamer variables in the question of dream recall. Sleep Research,
Butler, S. F., & Watson, R. (1985) Individual differences in memory
for dreams: The role of cognitive skills. Perceptual and Motor Skills,
Cohen, D. B. (1971) Dream recall and short-term memory. Perceptual
and Motor Skills, 33, 867-871.
Cory, T. L., Orniston, D. W., Simmel, E., & Dainoff, M. (1975)
Predicting the frequency of dream recall. Journal of Abnormal Psychology,
Gentil, M. L., & Lader, M. (1978) Dream contents and daydream attitudes
in anxious and calm women. Psychological Medicine, 8, 297-304.
Lloyd, S. R. (1976) Factors affecting dream recall: Visual memory,
repression, differentiation and rapid eye movements. Sleep Research, 5,
Rosenlicht, N., Maloney, T., & Feinberg, I. (1994) Dream report
length is more dependent on arousal level than prior REM duration. Brain
Research Bulletin, 34, 99-101.
Schredl, M. (1994) Dream length and creativity: An opposite finding.
Perceptual and Motor Skills, 78, 1994, 1297-1298.
Schredl, M. (1995a) Creativity and dream recall. Journal of Creative
Behavior, 29, 16-24.
Schredl, M., & Schröder, A. (1995b) Traumerleben von älteren
Menschen - Teil 2: Empirische Studie und Diskussion (Dreaming in the elderly
- Part II: empirical data and discussion). Zeitschrift für Gerontopsychologie
und -psychiatrie, (submitted).
Taub, J. M. (1972) Dream recall and content in long and short sleepers.
Perceptual and Motor Skills, 35, 267-270.
Waterman, D. (1991) Aging and memory for dreams. Perceptual and Motor
Skills, 73, 355-365.
Wood, J. M., Sebba, D., & Domino, G. (1989-90) Do creative people
have more bizarre dreams?: A reconsideration. Imagination, Cognition and
Personality, 9, 3-16.
Table 1: Means and standard deviations of the dream diary and memory
dream recall frequency .......... ...3.16.........1.48.................0...................
(mornings with dream recall)
dream report length .............. ..69.58.........68.38................1..................360
(mean word count)
subtest objects................... .....11.66...........
subtest city map................ ......21.22............6.27................5...................31
film sequence.................... .......31.90............6.96................21.................50