Available online at:
https://cc.arcabc.ca/islandora/object/cc%3Apsycjournal

J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.
Submitted Fall 2020; accepted July 2021.

What Is the Biological Cause of Dreams?
Authors: Chloe Nursey*, Devon Heathcote, and Kennedy Brewer
Supervising Instructor: Michael Pollock, Psyc 215 (“Biological Psychology”)
Department of Psychology, Camosun College, 3100 Foul Bay Road, Victoria, BC, Canada V8P 5J2
*Corresponding author email: chloenursey@outlook.com

ABSTRACT
The purpose of this experiment was to examine varying types of dreams and how they were
connected with our everyday lives. Previous research has shown that the biological causes of
dreams can be based on what type of sleep they are in, either REM or NREM, the amount of
cortical activation caused by what point of the circadian cycle the dreamer is in, and the type of
external or internal stimuli that may initiate the memory sources. In our first (correlational)
study, we tested the strength of these relationships by examining naturalistic daily changes in
their variables longitudinally over a two-week period. REM sleep was measured by using the
information provided from the phone app called Pillow, giving us the percentage spent in REM,
as well as the total minutes slept per night from each participant. Amount of sleep in the late
circadian morning was measured by recording how many hours past midnight each participant
slept in for each night of the study. The amount of light was measured with both the use of an
app called Light Meter upon waking each morning and the number of words used to describe
each recorded dream by their relevance to light stimulation. Dream recall was determined by the
average score that participants gave themselves on a scale of 0-10, depending on the clarity of
recalled dreams. Data pooled across participants in our correlational study showed no significant
correlations of REM sleep or sleep in the late circadian morning with dream recall. Similarly,
light stimulation showed no significant correlation with dream tallies. Although our correlations
were not significant, the strongest was sleep in the late circadian morning and dream recall.
Based on the strength of correlation found between the strength of dream recall and sleep in late
circadian morning in our correlational study, we then conducted a second (experimental) study to
test for specifically a causal relationship between these two variables. Over a ten-day period, we
randomly assigned participants each day to wake up either one hour earlier or at their normal
wake time and measured the effect this had upon dream recall each day. The results of our
experimental study showed statistical significance in pooled raw data but not in pooled
standardized data. The biological causes of dreams thus remain unsolved.
1. Introduction
1.1 Research Problem
There is much curiosity as to how our
minds are able to stimulate such varying

dreams and how they behave the way they
do. Our research goal is to gain insight as to
why some dreams can be remembered in
great detail, while other dreams may be hard
to remember once we wake up from our rest.
Having an understanding of what causes
131

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

dreams to occur may help those who wish to
alter their dreams, make sense of how their
dreams relate to their everyday lives, and
additionally, have comprehension of how
the connections between dreams and the
mind are formed. We wish to know what
biological mechanisms cause dreams so we
can have a better understanding for our own
sake, and learn how to make sense of why
we have specific dreams during sleep.
1.2 Literature Review
A factor as to why some dreams can be
remembered while others cannot, is based on
whether the person sleeping is engaged in
rapid eye movement (REM) or non-rapid
eye movement (NREM) sleep. A study was
conducted by Dement & Kleitman (1957) to
see what state of sleep would have a better
recall of the participants' dreams. The
participants consisted of nine adults, five
male, two females and they were told to
report to the laboratory a little before they
went to bed, they were also told to withhold
from consuming alcohol and caffeine
beverages on the day of the experiment. The
participants were recorded by an EEG
machine via scalp while they slept and were
woken multiple times during their sleep with
an alarm to see if they could recall their
dreams or not. In total, the participants slept
61 nights therefore there were a total of 152
dream recalls and 39 no recalls during REM
sleep, as well as 11 dream recalls and 149 no
recalls during NREM. This data shows that
80% of dreams were recalled during REM
sleep, while only 7% of dreams were
recalled during NREM sleep. This evidence
shows the logic behind why dreams are
sometimes remembered, but sometimes not
is based on whether the person is in REM or
NREM sleep before they are awakened.
Additionally, one theory suggests that
dreaming is produced by cortical activation

created by ultradian and circadian cycles.
Wamsley, Hirota, Tucker, Smith, and
Antrobus (2006) conducted an experiment to
try to better understand the way that
ultradian and circadian cycles influenced
dreaming. 20 undergraduate students in New
York were the participants for this study.
Each participant was recommended to keep
a sleep log 1 week before the experiment
officially began, in which during this week,
the participants were required to have an
average bedtime before midnight, which
could not be more than 1 hour away from
the set average, and keep a record of the
number of dreams that they were able to
recall each night, including notes on
remembered content in those dreams.
During the main experimental night,
participants went to bed 3 hours later than
the set time that they’d previously adhered
to during the week prior, and were also
encouraged to sleep in as long as possible on
the experimental night. Data was recorded
using electroencephalographic (EEG),
electrooculographic (EOG), and
electromyographic (EMG) montage
(Wamsley et al., 2006, p. 349). Each
participant was woken twice, once from a
REM and NREM period of their sleep near
the lowest point of their circadian body
temperature cycle, usually about 2.7 hours
before their average waking time the week
prior, and once from a REM and NREM
period at least 1 hour after what their wake
up time had initially, approximately when
their circadian-driven cortical activation
would be the highest. Awakenings during
REM sleep were done 10 minutes into a
period of REM, and NREM awakenings
were done during stage 2 of the participant’s
sleep, at least 20 minutes from the last time
that they were awake, or from the last period
of REM. Each participant was awoken when
their name was called on an intercom
speaker. Upon waking, the following
132

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

prerecorded questions were asked over the
intercom: 1. Tell me everything that was
going through your mind just before I called,
2. Tell me one more time everything that
was going through your mind just before I
called, 3. Were you trying to do anything?
(y/n), 4. Were you trying to find anything?
(y/n), 5. Were you trying to solve a
problem? (y/n), 6. Was anyone talking?
Could you hear any speech? (y/n).
(Wamsley et al., 2006, p. 349). After
responding to these questions, the
participants completed an assessment of the
emotional and visual qualities of what had
gone on in their minds before waking up,
including questions regarding visual clarity
and brightness of objects as well as
emotional intensity. These assessments were
then scored on a few different scales that
depicted their word information count,
dreamlike quality, and level of bizarreness.
Post conduction of the experiment, the
results showed that 80% of the participants
reported some form of mentation from the
lowest point in their circadian body
temperature cycle during REM period
awakening, 100% reported mentation during
the late REM period awakening, and 70%
reported mentation from both the lowest
body temperature point and late NREM
period awakenings. The researchers
concluded from this experiment that core
cognitive characteristics of the mentation
were most prominent during the late
morning in both REM and NREM, and that
although the cognitive features were
expressed more in REM reports than in
NREM reports, circadian-driven activation
amplified the expression similarly in reports
from both sleep stages (Wamsley et al.,
2006, p. 349). From the findings of this
study, it can be reasoned that a person’s
daily and circadian rhythm cycles have an
impact on when dreams are more likely to
occur, and that this influence can be applied

to dreaming in both REM and NREM
periods of sleep. Cortical activation is
assumed to be higher the further into the late
circadian morning that one sleeps in. This
suggests that someone would be more likely
to remember a dream if they had slept in
later than someone who had woken up early.
Another possible element to the cause of
dreams is the integration of sensory
stimulation. In a study conducted by Dement
& Wolpert (1958), twelve participants were
tasked with reporting to a laboratory shortly
before their usual sleep retirement, where
they would be connected to various
electrodes monitoring brainwave activity
and eye movements, and then left to fall
asleep in a darkened space. The researchers
utilized three different stimuli (a five-second
tone of 1000-hz, flashes of light and a light
spray of water from a syringe onto the
participants’ face) when Rapid Eye
Movement (REM) was present, and a bell
would be sounded to fully wake up the
participants and conclude the main testing
process. The results of this experiment
varied. Dreams transcripts, taken from tests
where participants did not awaken, included
hearing a close-by roaring sound akin to a
plane crash or earthquake, seeing shooting
stars or a flash of lightning, and building
leaks or rainfall. The water spray appeared
to have the most impact, with the 1000-hz
tone the least. In addition, the alarm bell,
despite not being one of the main stimuli,
was incorporated into dreams as they ended,
and were frequently described as a ringing
doorbell or telephone. The water spray,
while significantly lower in occurrence rate
comparatively, also bore a chance for similar
effect. A secondary study was conducted
subsequently, with Dement and Wolpert
instead putting focus toward the effect on
dreams made by stimuli from internal
processes. Three participants were tasked
with five instances of complete curtailment
133

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

of fluid consumption for a period of twentyfour hours minimum, and then to sleep in the
laboratory after each separate instance. All
participants, before falling asleep, reported
experiencing a great thirst, and symptoms
included a prominently dried mouth. The
results of this second experiment detailed
that while dream substance did not exhibit
particular awareness of thirst or the need
thereof, one-third of the dreams appeared to
indicate elements relevant to the imposed
restriction, including a toast being raised but
the participant having no cup of their own,
and the participant heating up roughly a litre
of milk for consumption. Based on these
results, it is possible that factors from
external and internal sources may play a role
in dream formation.
1.3 Hypotheses
Based on the above literature review, we
predicted the following hypotheses:
 Hypothesis #1: If REM increases during
sleep, then dream recall will increase.
 Hypothesis #2: If sleep occurs later in
the circadian morning, then dream recall
will increase.
 Hypothesis #3: If sensory stimulation
during sleep increases, then the quantity
of dream content relevant to that
stimulation will increase.
2. Methods
2.1 Participants
The three authors of this paper served as
the participants in its studies. The
participants ranged in age from 19-22 years
old, with an average age of 21 years, and
included three females. The participants
were all undergraduate students at Camosun
College who completed the current studies

as an assignment for Psyc 215 (“Biological
Psychology”) and were grouped together
due to their mutual interest in the biological
causes of dreams. All of the participants
reported having frequent dream sequences,
which fell within regular sleeping patterns.
2.2 Materials and Procedure
2.2.1 Correlational Study Methods
We first performed a correlational study
to test concurrently all of our hypotheses by
examining naturalistic daily changes in their
variables longitudinally. Each participant
kept a study journal with them at all times
over this study’s two-week period in order to
record self-observations of the following
four variables: (1) REM sleep, (2) Sleep in
late circadian morning, (3) sensory
stimulation during sleep, (4) dream content
relevant to sensory stimulation, and (5)
dream recall.
To measure the amount of time spent in
REM sleep, each participant recorded in
their study journals the percentage of REM
sleep they had throughout the night. This
information was provided by the Pillow app
and was calculated daily.
To measure the amount of sleep in the
late circadian morning, each participant
recorded how many hours that they slept
past midnight in a journal over the course of
the study. Using a chart of an assumed
timeline of circadian-driven cortical
activation during the night from the
Wamsley et al. (2006) article, participants
estimated that the level of cortical activation
would increase the later into the circadian
morning that they woke up (see appendix A
for the chart).
To measure sensory stimulation during
sleep, each participant recorded in their
study journals the level of light present in
their immediate surroundings, noted in Lux
through the app Light Meter, upon waking.
134

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

To measure dream content relevant to
sensory stimulation, dreams and their
content remembered were also written
down, and then passed towards a nonparticipant for review, where words
synonymous to the chosen element, such as
light and intensity, within each dream were
respectively tallied. The possible values on
this tally scale ranged from none (a score of
0), to moderate (a score of 5), to very high (a
score of 10). From these records, the average
tally scores for the dreams recorded by each
participated for each day was calculated.
To measure dream recall, participants
recorded in their study journals their ability
to remember their dreams on a 0 to 10 scale.
Clarity of dreams recollected was ranked
from 0 = no detail recalled, 5 = some detail
with partial obscurity, and 10 = full
recollection. From these records, the average
level of detail recalled in dreams of each
participant for each following day was
calculated.
To assess the strength and statistical
significance of associations between
variables predicted by our three hypotheses,
we performed Pearson product moment
correlations of their predictor variables
(REM sleep, level of a stimuli, and cortical
activation) with their outcome variable
(dream recall). For testing Hypothesis #1,
we correlated the total amount of time spent
in REM sleep each night with whether the
participant could recall their dream(s) in
detail when they woke up. For testing
Hypothesis #2, we correlated the level of
circadian-driven cortical activation with the
amount of detail in which participants could
remember the content of their dream(s). For
testing Hypothesis #3, we correlated the
nightly measure of stimulus-derived dream
content obtained by each participant for each
day. We performed all of the above
correlations separately for each participant
as well as using data pooled across all of the

participants. For the correlations using
pooled data, in addition to the raw data, we
also performed correlations after we had
first transformed the data from each
participant into z-scores in order to
standardize differences in averages and
variability seen between the participants in
their data and thus make them more
comparable. A correlation coefficient was
considered statistically significant if the
probability of its random occurrence (p) was
< .05 (i.e., less than 5% of the time expected
by chance alone).
2.2.2 Experimental Study Methods
Based on the strength of the strength of
the correlation between dream recall and
sleep in late circadian morning found in our
correlational study, we then chose to
conduct an experimental study to test for a
casual relationship between these two
variables for Hypothesis #2.
We manipulated the independent
variable, sleep in late circadian morning,
over a 10-day period by randomly assigning
participants each day to either an
experimental group, or control group. For
experimental days, participants woke up one
hour early from their normal wake time. For
control days, participants woke up at their
normal wake time. These wake times were
based on the average wake time participants
had over the 10-day period of the
correlational study. The wake times were
correspondent to each participant so no
external factors affected the study.
We were unable to make this a blind
procedure as participants needed to be aware
of what time they were waking each
morning to get to class and work on time. To
determine the conditions, the participants
were randomly assigned by flipping a coin
to either a control or experimental day
(control = heads, experimental = tails). On
experimental days, participants woke up one
135

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

hour earlier from their normal wake time.
On control days, participants would wake up
at their normal wake time. To control for
any bias resulting from rating their own
dream recall on an objective scale,
participants wrote their dream description in
a journal when they had waken, then had an
unbiased individual read their dream
description and rate it on the dream recall
scale (0 = could not remember any dreams,
10 = dreams could be remembered in
complete detail) depending on the level of
detail they thought was present.
To assess the statistical significance of
differences seen in dream recall on hourvariation wakeup experimental days vs.
regular wakeup control days, Student’s ttests were performed. We performed t-tests
separately for each participant as well as
using data pooled across all of the
participants. For the t-tests using pooled
data, in addition to using the raw data, we
also performed t-tests after we had first
transformed the data from each participant
into z-scores in order to standardize
differences in averages and variability seen
between the participants in their data and
thus make them more comparable. An
average difference between conditions was
considered statistically significant if, using a
two-tailed distribution (i.e., allowing this
difference to be positive or negative), the
probability of its random occurrence (p) was
<.05 (i.e., less than 5% of the time expected
by chance alone).
3. Results

morning was statistically stronger with
dream recall using both pooled raw data (r =
-.06, p = .728; see Figure 2.a) and pooled
standardized data (r = .20, p = .213).
Similarly, REM sleep and dream recall were
not significantly correlated for any single
participant, but showed similar strength to
sleep in late circadian morning using pooled
raw data (r = -.04, p = 0.147; see Figure 1.a)
and pooled standardized data (r = .13, p =
.413). In parallel, no statistically significant
correlations were found between light
stimulation and dream tallies using any
single participant’s data, pooled raw data (r
= .18, p = .256; see Figure 3.a), or pooled
standardized data (r = .15, p = .354),
however, one participant’s data briefly held
statistical significance. Based on a
comparison of the correlation coefficients
using either the pooled raw data or the
pooled standardized data, sleep in late
circadian morning showed the strongest
correlation with dream recall.
3.2 Experimental Study Results
As shown in Table 2, some significant
difference was found in dream recall
between early wake time and normal wake
time in the pooled raw data. Statistically
significant differences between these
conditions were not seen using pooled
standardized data (p = .123), however, any
single participant’s data (p ≥ .002) and
pooled raw data (p = .048 ; see Figure 4.a)
showed some evidence of significance, with
one participant’s data showing statistical
significance.

3.1 Correlational Study Results
4. Discussion
As shown in Table 1, both REM sleep
and sleep in late circadian morning were not
significantly correlated with dream recall.
Although not statistically significant for any
single participant, sleep in late circadian

4.1 Summary of Results
Based on previous research, we
hypothesized that increases in three
136

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

variables would be followed by greater
dream recall after each night of sleep: the
amount of REM sleep (Hypothesis #1), sleep
in late circadian morning, (Hypothesis #2),
and the amount of light stimulation
(Hypothesis #3). Data pooled across
participants in our correlational study
showed no significance between the
relationship of dream recall with REM sleep
and sleep in the late circadian morning
(Hypotheses #1&2). Additionally, no
significance was found between sensory
stimulation during sleep, and dream content
relevant to sensory stimulation (Hypotheses
#3). The strongest correlation was the
relationship between dream recall and sleep
in the late circadian morning. However, the
results of our experimental study did not
appear to establish a solid causal role of
sleep in the late circadian morning on dream
recall.
4.2 Relation of Results to Past Research
Our correlational study did not show a
significant relationship between REM sleep
and dream recall reported by previous
research. Dement & Kleitman (1957) found
that dream recall was based on whether the
participant was in REM sleep or NREM
sleep before they awakened. Dement &
Kleitman (1957) had nine adults participate
in their study, five males, and two females,
and were recorded by an EEG machine via
scalp while they slept and were asked to
recall their dreams when they woke up. In
contrast, our study had three female
participants, all in their twenties, and
recorded REM sleep through the Pillow app
and reported in study journals their ability to
recall dream(s), if any, on a 10-point scale. 0
meaning no recall of dream(s), and 10
meaning full recollection of dream(s). The
Pillow app was used to calculate the amount
of REM sleep the participants had each

night, calculated in percentage. Although
some data throughout the study suggests that
dream recall could be related to the amount
of REM sleep participants had, it was not
enough to show a significant relationship.
There was no significant relationship
found between sleep in the late circadian
morning and dream recall. This means that
our findings were not consistent with those
of the study conducted by Wamsely et al.
(2006). In the Wamsley et al. (2006) study
they found that waking up when levels of
cortical activation were assumed to be
higher, like later into the morning for
example, was correlated with higher dream
recall. While the Wamsely et al. (2006)
study used different technologies to measure
levels of cortical activation such as EEG,
EOG, and EMG, we used the chart included
in their article to estimate the level of our
own cortical activation at different hours of
the morning. We were able to use this chart
to assume that our cortical activation would
be higher in the later hours of the morning.
Similarly, while the Wamsley et al. (2006)
study included an assessment where the
participants ranked many factors such as
dreamlike quality, word information count,
and level of bizarreness, we chose to rank
our dream recall on a scale from 0-10. A 0
on this scale meant that no dream could be
remembered, and a 10 meant that a dream
could be remembered in complete detail.
Our study could not confirm that a
relationship between sleep in the late
circadian morning and dream recall does
exist. In our experimental study, we
restricted how late we would sleep into the
circadian morning by waking up an hour
earlier than we normally would. A limitation
we faced was our inability to conduct a
double-blind experiment, due to the fact that
alongside knowing the hour that we were
waking up at, our study design restricted late
circadian morning sleep by one hour in order
137

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

to go about our daily lives as usual. There
was no statistical significance found in our
experiment with the one hour difference
between conditions. It is recommended that
future studies use a larger gap between
conditions as well as a larger sample size to
find a significance between control and
experimental waking times.
Our correlational study failed to confirm
the relationship between sensory stimulation
and dream content relevance reported by
previous research, however, there was a
correspondence present. Dement & Wolpert
(1958) found that several aspects of
stimulation in sleep, measured through
equipment monitoring sleep activity and
transcripts of recounted dreams, appeared in
dreams within various formats, but dreams
themselves were not always present within
each testing period. Similarly, our
participants had dreams of varying strengths
and activity, while, however, experiencing
sleep periods where dreams involving the
chosen stimulation were either entirely
vague, and or incredibly undefined, or
nonexistent. The methodology of our
correlational study differed from that of
Dement & Wolpert (1958) in three major
ways that might account for the analogous
results. Foremost, differences between the
studies in how they measured dream content
relevance could have affected their findings.
Our correlational study relied only upon
self-reports of synonymous words used in
dream transcripts. Future studies should test
whether the objectively verified aspects of
stimulation during sleep outlined by Dement
& Wolpert (1958), but not subjective selfassessments, predict dream content severity.
Second, differences between the studies in
how they measured content relevance in
dreams could have affected their findings.
Lending credence to this possibility is the
fact that Dement & Wolpert (1958) could
not entirely predict the strength and effects

of a sensory stimulus on dreams, such as
how strongly an element based around the
utilized stimulation was present, and the
perceived time duration within each dream.
While we measured content relevance based
on a tally scale, Dement & Wolpert (1958)
based their results on described events in
dream transcripts. Future studies should
examine whether sensory stimulation is able
to predict the severity of only certain selfreported content-relevant dream transcripts.
Lastly, differences between the studies in
how sensory stimulation was received could
have affected their findings. Although
testing conditions could be manipulated to
the experimenter’s desirability, there would
be little to no accounting for stimulus
reception prior to and post the
experimentation period. Future studies
should examine the exposure levels of
stimulation received, separate and prior to
those of manipulated circumstances.
4.3 Implications of Results
Possible practical applications of our
current findings are the increased memory of
dream content. For example, this could
prove to be beneficial toward the creation of
innovative ideas appearing to someone in a
dream, and they wish to remember them for
later application to their lives, but not when
or how the innovations would potentially be
administered. Additionally, as dreams are
often noted as a subconscious means of
processing data from both recent and past
experiences, an improved recollection of
dreams could provide a window for curious
minds, whether by the review of prior events
or pure introspection, and potentially include
this possibility for future research.
We originally conducted the current
studies to find a way to better recall content
details within dreams that would otherwise
be lost normally. Unfortunately, based on
138

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

our experimental results, we cannot
accurately define a method where sleeping
later into the circadian morning would
ensure higher dream recall. It remains up to
future studies to find a working solution to
this problem however possible.

References
Dement, W., & Kleitman, N. (1957). The
relation of eye movements during sleep to
dream activity: An objective method for
the study of dreaming. Journal of

Experimental Psychology, 53(5), 339346.
Dement, W., & Wolpert, E. A. (1958). The
relation of eye movements, body motility,
and external stimuli to dream content.
Journal of Experimental Psychology,
55(6), 543–553. https://doiorg.libsecure.camosun.bc.ca:2443/10.103
7/h0040031
Wamsley, E. J., Hirota, Y., Tucker, M. A.,
Smith, M. R., Antrobus, J. S. (2006).
Circadian and ultradian influences on
dreaming: A dual rhythm model. Brain
Research Bulletin, 71(4), 347-354.

139

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

Table 1
Correlation coefficient (r) values, with the number of daily trials (n) per correlation in brackets.
Variables
correlated

Participant
#1

Participant #2

Participant
#3

Pooled
raw data

Pooled
standardized
data

REM sleep &
dream recall

0.41(14)

-0.15(14)

0.13(14)*

-0.04(42)

0.13(42)

Sleep in late
circadian morning
& dream recall

0.49(14)

-0.21(14)

0.31(14)

-0.06(42)

0.20(42)

Light stimulation
& dream tallies

0.27(14)

-0.33(14)

0.51(14)

0.18(42)

0.15(42)

* p < .05.

140

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

Table 2
Descriptive statistics on dream recall for early wake time condition and normal wake time
condition.

Condition

Statistic

Participant Participant Participant
#1
#2
#3

Pooled
raw
data

Pooled
standardized
data

Early Wake
Time
(Experimental)

Normal Wake
Time (Control)

Mean

3.33

5.50

2.71

3.79*

-0.22

S.D.

3.44

2.17

3.73

3.28

0.86

n

6

6

7

19

19

Mean

9.50

4.00

6.33

6.64

0.38

S.D.

0.58

3.74

4.73

3.85

1.06

n

4

4

3

11

11

Dream recall was measured on a 0-10 scale of clarity, where 0 = no details remembered and 10 =
full detail remembered
*p < .05 for comparison of early wake time condition with its respective normal wake time
condition.

141

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

Figure 1.a
Scatterplot of REM sleep and dream recall using pooled raw data across participants.

Figure 1.b
Scatterplot of REM sleep and dream recall using pooled standardized data across participants.

Marker color indicates which participant data is from: red = participant #1, orange = participant
#2, and yellow = participant #3. Some data might not be visible in the figure due to overlapping
markers.
142

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

Figure 2.a
Scatterplot of sleep in late circadian morning and dream recall using pooled raw data across
participants.

Figure 2.b
Scatterplot of sleep in late circadian morning and dream recall using pooled standardized data
across participants.

Marker color indicates which participant data is from: red = participant #1, orange = participant
#2, and yellow = participant #3. Some data might not be visible in the figure due to overlapping
markers.
143

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

Figure 3.a
Scatterplot of light stimulation and dream tallies using pooled raw data across participants.

Figure3.b
Scatterplot of light stimulation and dream tallies using pooled standardized data across
participants.

Marker color indicates which participant data is from: red = participant #1, orange = participant
#2, and yellow = participant #3. Some data might not be visible in the figure due to overlapping
markers.
144

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

Figure 4.a
Bar graph of dream recall across early wake time and normal wake time conditions using pooled
raw data from participants, with error bars showing ± 95% confidence levels, and with an
overlapping scatterplot of data from each participant.

145

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

Figure 4.b
Bar graph of dream recall across early wake time and normal wake time conditions using pooled
standardized data from participants, with error bars showing ± 95% confidence levels, and with
an overlapping scatterplot of data from each participant.

Marker indicates which participant data is from: red - participant #1, orange = participant #2, and
yellow = participant #3.

146

Nursey, Heathcote, & Brewer - J Camosun Psyc Res. (2021). Vol. 3(1), pp. 131-147.

Appendix A

Chart depicting assumed timeline of level circadian-driven cortical activation throughout the
night.

147