Reality Testing and the Mnemonic Induction of Lucid Dreams: Findings From the National Australian Lucid Dream Induction Study

Abstract

Lucid dreaming is a learnable skill and has a wide range of potential applications. However, research in this area has been limited by a lack of effective and reliable lucid dream induction techniques. The present study provides a thorough investigation into 3 of the most promising cognitive lucid dream induction techniques-reality testing, wake back to bed (WBTB), and the mnemonic induction of lucid dreams (MILD) technique. A sample of 169 Australian participants completed a pretest questionnaire, provided baseline logbook data in Week 1, and practiced lucid dream induction techniques in Week 2. Results showed that the combination of reality testing, WBTB and the MILD technique was effective at inducing lucid dreams. Several factors that influenced the effectiveness of the MILD technique were identified, including general dream recall and the amount of time taken to fall asleep after finishing the technique. Recommendations for future research on lucid dream induction are provided.

Full text

Reality Testing and the Mnemonic Induction of Lucid
Dreams: Findings From the National Australian Lucid
Dream Induction Study
Denholm J. Aspy, Paul Delfabbro, Michael Proeve, and Philip Mohr
The University of Adelaide
Lucid dreaming is a learnable skill and has a wide range of potential
applications. However, research in this area has been limited by a lack of effective
and reliable lucid dream induction techniques. The present study provides a thorough
investigation into 3 of the most promising cognitive lucid dream induction
techniques—reality testing, wake back to bed (WBTB), and the mnemonic induction
of lucid dreams (MILD) technique. A sample of 169 Australian participants
completed a pretest questionnaire, provided baseline logbook data in Week 1, and
practiced lucid dream induction techniques in Week 2. Results showed that the
combination of reality testing, WBTB and the MILD technique was effective at
inducing lucid dreams. Several factors that influenced the effectiveness of the MILD
technique were identified, including general dream recall and the amount of time
taken to fall asleep after finishing the technique. Recommendations for future
research on lucid dream induction are provided.
Keywords: lucid dreaming, induction techniques, dream recall, methodology
A lucid dream is a dream in which the dreamer is aware that they are dreaming
while the dream is still happening (LaBerge, 1985). In a recently published
meta-analysis, Saunders, Roe, Smith, and Clegg (2016) found that an estimated
55% of adults have experienced at least one lucid dream in their lives, with 23% of
adults experiencing lucid dreaming regularly (once per month or more) and some
rare individuals having spontaneous lucid dreams almost every night (Fingerlin,
2013;Gackenbach, 1991;Schredl & Erlacher, 2011b;Snyder & Gackenbach, 1988).
Although references to lucid dreaming can be found from over 2000 years ago
(LaBerge, 1985), it was not until 1975 that the phenomenon was confirmed
empirically. Hearne (1978) reasoned that, if the eye movements that characterize
REM sleep correspond to the gaze of the dreamer, it may be possible to signal to
the outside world during a lucid dream using a series of prearranged left-right eye
Denholm J. Aspy, Paul Delfabbro, Michael Proeve, and Philip Mohr, School of Psychology, The
University of Adelaide.
We thank Thomas McAtee for his valuable assistance with data entry and data management.
Correspondence concerning this article should be addressed to Denholm J. Aspy, School of
Psychology, The University of Adelaide, North Terrace Campus SA 5005, Australia. E-mail: denholm
.aspy@adelaide.edu.au
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
206
Dreaming © 2017 American Psychological Association
2017, Vol. 27, No. 3, 206–231 1053-0797/17/$12.00 http://dx.doi.org/10.1037/drm0000059

movements. Using electrooculography, he succeeded in recording such a signal that
corresponded to the report of a lucid dream during unambiguous REM sleep. This
was achieved independently by LaBerge (1980), and numerous other studies have
since replicated these findings (e.g., Dane, 1984;Fenwick et al., 1984;Ogilvie, Hunt,
Tyson, Lucescu, & Jeakins, 1982;Tholey, 1983). There was a subsequent surge of
research into such topics as the phenomenology (see LaBerge & DeGracia, 2000),
psychophysiology (see Schredl & Erlacher, 2011a), and potential applications of
lucid dreaming. Potential applications include treatment of nightmares (Holzinger,
Klösch, & Saletu, 2015;Lancee, van den Bout, & Spoormaker, 2010;Spoormaker
& van den Bout, 2006), improvement of physical skills and abilities through
rehearsal in the lucid dream environment (Erlacher & Schredl, 2010;Stumbrys,
Erlacher, & Schredl, 2016), creative problem solving (Stumbrys & Daniels, 2010),
and research opportunities for exploring consciousness and mind-body relation-
ships (see Hobson, 2009). However, research on lucid dreaming has been limited by
a lack of effective and reliable lucid dream induction techniques.
Lucid Dream Induction Techniques
A wide range of techniques have been developed for inducing lucid dreams
(see LaBerge & Rheingold, 1991;Love, 2013;Stumbrys, Erlacher, Schädlich, &
Schredl, 2012;Tholey, 1983). Some of these involve dream induced lucid dreams
(DILDs), which occur when the dreamer realizes they are dreaming during a
nonlucid dream (LaBerge & Rheingold, 1991). Other techniques are designed to
help the practitioner enter a lucid dream directly from the waking state. These are
known as wake induced lucid dreams (WILDs; LaBerge & Rheingold, 1991) and
require a delicate balance of relaxation and unbroken conscious awareness during
the transition into REM sleep. WILDs are considered more difficult to achieve, and
WILD attempts carry an increased risk of experiencing sleep paralysis, which can
be extremely unpleasant (Cheyne, 2003;Sharpless & Barber, 2011). DILD tech-
niques are thus considered better suited to beginners and have been favored in
lucid dream induction research. Stumbrys et al. (2012) further organized lucid
dream induction techniques according to three broad categories. Cognitive tech-
niques include all techniques that involve cognitive activities for inducing lucid
dreams. These techniques do not require specialized equipment and are the most
widely used for inducing lucid dreams. External stimulation techniques involve the
presentation of stimuli such as flashing lights, acoustic stimuli, and mild electric
shocks during REM sleep, which can be incorporated into the dream experience
and serve as a cue to the dreamer that they are dreaming. Miscellaneous techniques
cover techniques that do not fit into the other two categories such as the ingestion
of specific substances that may promote lucid dreams (see LaBerge, 2004; see also
Yuschak, 2006).
The two most widely studied cognitive techniques for inducing DILDs and the
subjects of the present study are reality testing (LaBerge & Rheingold, 1991;
Tholey, 1983) and the Mnemonic Induction of Lucid Dreams (MILD) technique
(LaBerge, 1980;LaBerge & Rheingold, 1991). Reality testing involves examining
one’s surroundings multiple times throughout the day, questioning whether one is
awake or dreaming, and then performing a reliable reality test to determine whether
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 207

one is awake or dreaming. Reality testing is important because of the strong
tendency for the dreaming mind to explain away even the most obvious indicators
that one is dreaming. One of the most popular reality tests involves rereading
written text (written text tends to change upon second inspection in dreams;
LaBerge & Rheingold, 1991). However, written text may not always be available,
and some lucid dreamers favor the inhalation test, which involves closing one’s lips
and then attempting to inhale (see Love, 2013). Performing this action while
dreaming can produce a contradictory sensation of air moving through one’s closed
mouth (presumably, this is because the muscles of the face and mouth but not those
involved in respiration are inhibited during REM sleep). The rationale behind
reality testing is that if it becomes habitual it will eventually be performed while
dreaming, thereby leading to lucidity.
The MILD technique makes use of prospective memory, which is the ability to
remember to perform planned actions in the future. While lying in bed and
immediately prior to going to sleep, the practitioner repeats the phrase “next time
I’m dreaming, I will remember that I’m dreaming” (or some variation) while
imagining themselves becoming lucid in a dream. If successful, this intention will be
remembered during subsequent REM sleep, and the dreamer will become lucid.
The MILD technique is often combined with another lucid dream induction
technique known as Wake Back to Bed (WBTB; LaBerge & Rheingold, 1991). This
involves waking up after several hours of sleep (usually five to six hours) and
remaining awake for a period of time (from as little as 10 min to more than one
hour) before returning to sleep. WBTB not only increases mental alertness, but also
provides an ideal time to practice the MILD technique. This is because REM sleep
is entered more quickly and for longer periods as sleep progresses, and most dreams
(including lucid dreams) occur during REM sleep (Nielsen, 2000;Stumbrys &
Erlacher, 2012). Thus, the intention to remember that one is dreaming is more
likely to be retained during REM sleep if the MILD technique is practiced after five
to six hours of sleep.
Research on Lucid Dream Induction
Stumbrys et al. (2012) identified a total of 35 empirical studies in a recent
systematic review of the lucid dream induction literature. Of these, 11 were
conducted in sleep laboratories and 24 were field studies. Most studies (27)
investigated cognitive techniques, with the majority (22) being field studies. A total
of 10 studies investigated the MILD technique. One was a sleep laboratory study
(Kueny, 1985), and the others were field studies conducted by LaBerge, Levitan
and their colleagues (Edelstein & LaBerge, 1992;LaBerge, 1988;LaBerge, Phillips,
& Levitan, 1994;Levitan, 1989,1990a,1990b,1991;Levitan & LaBerge, 1994;
Levitan, LaBerge, & Dole, 1992). Reality testing was investigated in nine studies.
One of these was a sleep laboratory study (Dane, 1984), with the others being
conducted in-field (LaBerge, 1988;Levitan, 1989;Levitan & LaBerge, 1994;
Malamud, 1979;Purcell, 1988;Purcell, Mullington, Moffitt, Hoffmann, & Pigeau,
1986;Reis, 1989;Schlag-Gies, 1992). An additional field study investigating reality
testing has been recently published (Taitz, 2011). Stumbrys et al. (2012) concluded
that MILD and reality testing appear to be more effective than most other cognitive
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
208 ASPY, DELFABBRO, PROEVE, AND MOHR

techniques. However, one study found that reality testing did not increase lucid
dreaming frequency (LaBerge, 1988), as did the more recent study by Taitz (2011).
It is unclear how reality testing compares to MILD. One study found that reality
testing was more effective at inducing lucid dreams (Levitan, 1989), but another
found that MILD was superior (LaBerge, 1988). Other than the study by Taitz
(2011), only two other lucid dream induction studies have been published following
the review by Stumbrys et al. (2012). One of these investigated visual (flashing
lights) and tactile (vibration) external stimulation techniques (Paul, Schadlich, &
Erlacher, 2014), and the other applied transcranial DC stimulation (tDCS) to the
dorsolateral prefrontal cortex (DLPFC) during REM sleep (Stumbrys, Erlacher, &
Schredl, 2013b). Both of these studies reported poor success rates.
Unfortunately, it is difficult to compare the effectiveness of lucid dream
induction techniques across studies due to widespread methodological limitations.
Stumbrys et al. (2012) evaluated the studies included in their review using a
methodological quality checklist, developed by Downs and Black (1998) that
assesses quality of reporting, external and internal validity, and statistical power.
Most (60%) studies were classified as poor quality, with the rest (40%) classified as
moderate quality. Just over half of the studies were either unpublished doctoral
dissertations or were otherwise not published in academic journals. The average
score for all studies was 9.1 out of a maximum of 28, with no substantial difference
between laboratory and field studies. Field studies investigating reality testing
scored slightly higher at 11.5. Field studies investigating the MILD technique
scored below average at 5.9 and were all published in a nonacademic magazine
targeted to lucid dreaming enthusiasts. All 35 studies included in the review scored
poorly on external validity, with participants mostly consisting of self-selected lucid
dream enthusiasts or university students. Other common issues included insufficient
statistical power (because of small sample sizes), lack of random allocation, and
invalid or unreliable outcome measures.
Inconsistent operationalization of lucid dreaming rates is another problem in
the empirical lucid dream induction literature, as well as the broader empirical
literature on general dream recall (see Aspy, 2016;Aspy, Delfabbro, & Proeve,
2015). A common operationalization is the mean number of lucid dreams reported
in a given period (dream count; DC). However, lucidity is often lost and regained
within a single dream and may be attained in multiple separate dreams, especially
among proficient lucid dreamers. This makes it hard to compare studies of
proficient lucid dreamers with studies involving less proficient participants. An-
other operationalization is the percentage of all reported dreams that are lucid. This
suffers the same limitations as DC but is also confounded with general (nonlucid)
dream recall rates. The percentage of participants that experience lucid dreaming
at least once while trialing a technique avoids the aforementioned problems, but is
too insensitive to provide much insight into technique effectiveness on its own. A
more suitable operationalization is lucid dream recall frequency (DRF; see Aspy et
al., 2015), which is the proportion of days in a given period on which lucid dreaming
is experienced irrespective of how many lucid dreams are reported. Mean DRF
rates are less prone to being inflated by participants who have multiple lucid dreams
within the same night or who are more likely to lose and then regain lucidity in the
same dream, thus making it easier to compare findings from studies that have
different sample characteristics.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 209

A further psychometric consideration is that dream recall can be measured using
both retrospective measures (which involve estimating one’s dream recall for a recent
specified time period such as the past month) and logbook measures (which involve
keeping a daily record of one’s dream recall). As discussed by Aspy (2016; see also
Aspy, Delfabbro, & Proeve, 2015), logbook measures of dream recall tend to have
higher internal validity and are more sensitive to daily fluctuations of dream recall, but
may have lower ecological validity due to their tendency to enhance dream recall. This
is because logbook measures require participants to spend more time thinking about
and retrieving dream memories each morning. Retrospective measures avoid this
enhancement effect and thus have higher ecological validity, but have lower internal
validity due to their tendency to underestimate dream recall rates. An example of a
retrospective measure of lucid dream recall frequency is the eight-point scale devel-
oped by Schredl and Erlacher (2004; see also Stumbrys, Erlacher, & Schredl, 2013a),
which includes response options such as “Several times a week” and “About once a
week.”
One of the biggest limitations in the empirical lucid dream induction literature is
the near-ubiquitous failure to measure variables related to how lucid dream induction
techniques were practiced. Only one study (LaBerge, 1988), which investigated both
MILD and reality testing in different experimental groups, reported correlations
between the number of technique repetitions and lucid dreaming rates. This study
found that the number of times the MILD phrase was repeated each night was
correlated with lucid dreaming (r⫽.12). In contrast, the correlation between lucid
dreaming and the number of reality tests performed was very small and nonsignificant
(r⫽.04). However, it is difficult to interpret these findings because essential statistical
and methodological information was not reported. More recently, Taitz (2011)
reported having measured the number of reality tests performed by participants but
provided no descriptive statistics, and this variable was not included in analyses. To
properly assess lucid dream induction techniques and maximize their effectiveness, it is
essential that variables related to technique practice are examined in addition to overall
lucid dreaming rates.
Aims and Hypotheses
The present study forms part of a larger research project that also
investigated issues related to the measurement of dream recall (see Aspy, 2016).
The aim of the present study was to provide a thorough investigation into reality
testing, WBTB, and the MILD technique. Baseline logbook data were collected
during Week 1 of the study and then participants were randomly allocated to
one of three experimental groups for Week 2. Because reality testing, WBTB
and MILD are often used in combination, and in the interests of identifying a
maximally effective approach to lucid dream induction, groups involving reality
testing only (reaction time (RT) only group), reality testing and WBTB (RT ⫹
WBTB group) and reality testing, WBTB and MILD (RT ⫹WBTB ⫹MILD
group) were compared. The RT ⫹WBTB condition involved reading a
document about lucid dreaming (see Section RT WBTB group), which con-
trolled for the effects of thinking about lucid dreaming that are inherent to the
MILD technique.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
210 ASPY, DELFABBRO, PROEVE, AND MOHR

Hypotheses were as follows:
●Because the relationship between lucid dreaming and general dream recall rates
is one of the most robust relationships observed in the empirical lucid dreaming
literature (see Erlacher, Schädlich, Stumbrys, & Schredl, 2014), it was hypoth-
esized that there would be significant positive correlations between general
dream recall rates and lucid dreaming rates at both pretest and during Week 2.
●It was hypothesized that lucid dreaming rates would be significantly higher
in Week 2 compared with Week 1 for all participants combined and for
participants in each of the three Week 2 groups.
Exploratory analyses were also conducted to investigate relationships between
lucid dreaming rates and a range of other variables described in Section Materials,
including variables that operationalize the way in which lucid dream induction
techniques were practiced.
Method
Participants
A total of 420 participants who passed the exclusion criteria (below) signed up for
the study and completed the pretest questionnaire. A total of 169 participants went on
to complete the full study. This final sample consisted of 94 (55.6%) females, 73
(43.2%) males and 2 (1.2%) participants who identified their gender as “other.” The
mean age was 38.3 (SD ⫽15.0) and ranged from 18 to 75. Most of the participants were
employed nonstudents (n⫽116, 68.6%), with 36 (21.3%) participants being students
and 17 (10.1%) being unemployed or retired. Most participants (63.9%) had no prior
experience with lucid dream induction techniques. Participants in the final sample
heard about the study from a range of recruitment sources: 54 (32.0%) from physical
posters or flyers distributed in public locations across the Australian states of South
Australia, Victoria, and New South Wales; 32 (18.9%) from word of mouth; 24 (14.2%)
from nationally televised news interviews with the first author; 23 (13.6%) from
newspaper articles; 15 (8.9%) from radio interviews; 12 (7.1%) from social media; and
9 (5.3%) from other Internet sources. Participants were excluded from the study if they
had been diagnosed with any kind of mental health disorder, sleep disorder, or
neurological disorder; suspected they might have one of these disorders; were experi-
encing a traumatic or highly stressful life event that was interfering with their sleep;
suffered from persistent insomnia or were unable to keep a regular sleep schedule; had
experienced sleep paralysis more than once in the past 6 months; found it unpleasant
to think about their dreams; or were under 18 years of age. All participants who
completed the study entered a raffle to win one of five $200 gift vouchers or one of 10
$50 gift vouchers.
Materials
Materials included an online pretest questionnaire and physical packages that
contained an instructions sheet, Week 1 logbook, and a sealed envelope containing
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 211

materials for Week 2. This envelope had the words “Week 2 materials—do not
open until Week 1 is complete” printed on the front to discourage participants from
attempting the lucid dream induction techniques prematurely. All participants
reported that they complied with these instructions. The Week 2 envelopes
contained another instructions sheet, lucid dream induction technique documents,
and a Week 2 logbook. Some of these materials are described in greater detail by
Aspy (2016). In the present paper, pretest variables are identified by a capital “P”
and logbook variables by a capital “L.”
Pretest questionnaire. The pretest questionnaire included the following
measures:
Demographic questions. Participants were asked to indicate their age, gen-
der, occupation, and how they heard about the study.
General dream recall. Two retrospective measures of general dream recall
were used. The first assessed DRF over the last week (PDRF; the percentage of
days on which there was dream recall) by asking “How many days during the last
week did you remember your dreams from the previous night?” Participants
selected one of eight options from a drop-down menu ranging from “0 days” to “7
days.” Following this, the number of separate dreams recalled over the past week
was assessed by asking “On average, how many separate dreams do you usually
remember per week?” Participants could select any whole number between 0 and
50 or “more than 50” from a drop-down menu. The mean number of dreams
recalled per day (P DC per day) was attained by dividing responses by seven.
Lucid dream recall. A question adapted from Brown and Donderi’s (1986)
Sleep and Dream Questionnaire (SDQ) assessed retrospective DC dream recall for
lucid dreams (P DC lucid per month):
Lucid dreams are those in which a person becomes aware of the fact that he or she is
dreaming while the dream is still ongoing. For example: ‘I was in England talking to my
grandfather when I remembered that (in real life) he had died several years ago and that I
had never been to England. I concluded that I was dreaming and decided to fly to get a bird’s
eye view of the countryside. . .’ Please estimate the number of lucid dreams you have had in
the past month.
Participants answered by selecting any whole number from 0 to 30 or “more than
30” from a drop-down menu. A limitation of this measure is that it does not capture
variation among infrequent lucid dreamers (⬍once per month). However, this
measure was chosen in favor of Schredl and Erlacher’s (2004) eight-point lucid
dreaming scale due to concerns that their scale would not capture variance among
frequent lucid dreamers (the most frequent response option in their scale is
“Several times a week”).
Prior lucid dream induction technique practice. Participants were asked
“Have you ever tried to have lucid dreams by learning and then practicing a lucid
dreaming technique?” (P lucid tech prior; “Yes” or “No”). Participants were then
asked “How often have you practiced a lucid dreaming technique recently (in the
past several months)?” (P lucid tech freq). Response options from Schredl’s (2004)
widely used dream recall measure were used (0 ⫽never;1⫽less than once a month;
2⫽about once a month;3⫽two or three times a month;4⫽about once a week;
5⫽several times a week; and 6 ⫽almost every morning). Responses were converted
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
212 ASPY, DELFABBRO, PROEVE, AND MOHR

to the approximate number of days per week using the following class means: 0 ⫽
0;1⫽0.125; 2 ⫽0.25; 3 ⫽0.625; 4 ⫽1.0; 5 ⫽3.5; 6 ⫽6.5.
Logbooks. Three different Week 1 logbooks were used. They each used a
different primary measure of general dream recall but were otherwise identical.
The Checklist logbook elicited brief titles for each dream recalled. The Narrative
logbook elicited detailed written narratives of each dream recalled. The Quantity
logbook prompted participants to rate the extent to which each dream was recalled
(see Section General dream recall). Aspy (2016) provided a detailed comparison of
these three types of logbooks and found that measures of general dream recall were
most stable and reliable when presented in the Quantity logbook, with no
differences in overall dream recall rates. These findings were anticipated prior to
conducting the study, and the Quantity logbook was used in all three of the Week
2 groups (with additional questions related to lucid dream induction techniques
described in Section Sleep-related questions).
Preliminary questions. Participants indicated the date of each logbook entry,
allowing the number of days taken to complete all seven entries to be calculated.
The total number of logbook entries made by each participant was also counted.
General dream recall. Participants were asked if they could recall anything
specific about their dreams from the previous night and were asked to provide brief
titles for each dream recalled. This allowed dream recall to be operationalized as
both dream recall frequency (LDRF; the percentage of days on which there was
dream recall) and dream count (L DC per day; the number of dreams recalled each
day). Participants were also asked to rate the amount of content recalled from each
dream using four categories provided. This operationalization is referred to as
dream quantity (LDQ) and was developed by Aspy (2016) based on an earlier
measure developed by Reed (1973). Category ratings are converted to numerical
values (“Fragmentary” ⫽1, “Partial” ⫽2, “Majority” ⫽4, “Whole” ⫽8) and
summed (higher scores indicate superior dream recall).
Lucid dream recall. Lucid dreaming was operationalized as DRF (the
percentage of mornings when lucid dreaming was reported) because in many cases
participants were unsure of how many lucid dreams they had, and in some cases lost
lucidity and then regained it within the same dream. The following question was
used: “Did you have any lucid dreams last night? (Lucid dreams are those in which
a person becomes aware of the fact that he or she is dreaming while the dream is
still ongoing)” (“yes” or “no”; L DRF lucid). Days when participants did not
practice lucid dream induction techniques were excluded when calculating Week 2
L DRF lucid rates. The percentage of participants that experienced lucid dreaming
at least once during Week 1 and during Week 2 was included as a second
operationalization of lucid dreaming (L lucid participants). Participants were also
asked “How long (approximately) do you think you were lucid dreaming?...
minutes” (L lucid duration min).
Sleep-related questions. Participants were asked to estimate how much time
they had spent sleeping (L time asleep): “How much time in total do you think you
spent sleeping last night?...hours,...minutes.” Participants also rated their
subjective sleep quality (L sleep quality): “On a scale of 1 to 5, what was the overall
quality of your sleep last night?” (1 ⫽terrible;2⫽poor;3⫽okay;4⫽good;5⫽
excellent). Participants indicated how tired they felt upon waking (L tiredness on
waking) with the following question: “On a scale of 1 to 5, how tired do you feel this
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 213

morning?” (1 ⫽not at all tired;2⫽slightly tired;3⫽somewhat tired;4⫽quite tired;
5⫽very tired). Finally, participants indicated how sleep deprived they were the
previous day (L sleep dep yesterday): “On a scale of 1 to 5, how sleep deprived were
you yesterday?” (1 ⫽not at all;2⫽slightly;3⫽somewhat;4⫽quite;5⫽very).
Lucid dream induction technique practice questions. The following question
was used in all three Week 2 logbooks: “How many reality tests did you perform
yesterday?” (blank space provided; L reality tests). In the RT ⫹WBTB and RT ⫹
WBTB ⫹MILD groups, the following additional questions were included: “Were
you in the middle of a dream when the alarm woke you up to do the technique?
(“yes,” “no,” or “unsure”; L awoke while dreaming); “On a scale of 1 to 5, how
motivated did you feel about doing the technique after the alarm went off?” (1 ⫽
not at all motivated;2⫽slightly motivated;3⫽somewhat motivated;4⫽quite
motivated;5⫽very motivated;L technique motivation); “On a scale of 1 to 5, how
difficult was it to focus on the technique?” (1 ⫽not at all difficult;2⫽slightly
difficult;3⫽somewhat difficult;4⫽quite difficult;5⫽very difficult;L difficulty
focusing); “How long (approximately) did it take for you to get to sleep after you
did the technique?...minutes.” (L min back to sleep). The following questions
specific to the MILD technique were included for participants in the RT ⫹
WBTB ⫹MILD group: “Did you fall asleep while you were still trying to do the
technique?” (“yes” or “no”; L asleep during technique); “If you answered ‘no’ to
the above question, how long (approximately) did it take for you to get to sleep
after you stopped doing the technique?...minutes.” (L min back to sleep); “How
long (approximately) did you spend on doing the technique?...minutes.” (L
technique min); and “How many times (approx) did you repeat ‘next time I’m
dreaming, I will remember I’m dreaming’?” (blank space provided; L technique
repetitions).
Lucid dream induction technique documents. Participants were given differ-
ent lucid dream induction technique documents depending on which of the Week
2 groups they were in.
RT only group. The “daytime lucid dreaming technique” document in-
structed participants to perform a minimum of 10 reality tests per day by first asking
themselves “Am I dreaming?” The importance of genuinely considering the
possibility that they are dreaming was emphasized. They were instructed to
examine their surroundings for anything strange or inconsistent. They were then
instructed to perform an inhalation reality test (see Section Lucid Dream Induction
Techniques). Participants were asked to count the number of reality tests per-
formed each day using one of several free tally counter apps available for iPhone
and Android smartphones or by making marks on a piece of paper or the back of
their hand. Participants were told that reality testing is most effective when
practiced frequently and carefully, and that reality tests should be performed at a
range of times and settings throughout the day (especially when something unusual
or unexpected happens).
RT ⴙWBTB group. In addition to the “daytime lucid dreaming technique”
document, these participants were given a “nighttime lucid dreaming technique”
document that outlined the WBTB technique. It instructed participants to set an
alarm for five hours after going to bed and place it somewhere where they would
have to get out of bed to turn it off. They were instructed to put a light on when
their alarm went off, go to the bathroom if necessary, return to bed and then read
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
214 ASPY, DELFABBRO, PROEVE, AND MOHR

a document entitled “what to do if you have a lucid dream” before returning to
sleep as they normally would. This document—which was given to participants in
all three Week 2 groups—was approximately 700 words long and began by
explaining several ways that lucid dreams can happen. It advised participants that
if they became lucid, they should stay calm to avoid waking, perform a reality test,
then stabilize the dream by rubbing the palms of their hands together vigorously
and focusing on the physical sensations while repeating “this is a lucid dream”
(LaBerge & Rheingold, 1991). Participants were asked to perform a reality test
upon reaching the end of the document.
RT ⴙWBTB ⴙMILD group. Participants in this group were given the
“daytime lucid dreaming technique” and “what to do if you have a lucid dream”
documents, as well as a “nighttime lucid dreaming technique” document that
outlined the MILD technique (LaBerge, 1980;LaBerge & Rheingold, 1991).
Participants were instructed to set an alarm for five hours after going to bed, put
a light on when their alarm went off and then sit upright in bed and try to
remember a dream from just before they woke up (or any recent dream if they
were unable to recall one). They were then told to go to the bathroom if
necessary before turning off the light, lying down comfortably and repeating the
phrase “next time I’m dreaming, I will remember that I’m dreaming.” The
importance of putting meaning into the words was emphasized. Participants
were told to simultaneously imagine themselves back in the dream they had
recalled and noticing something unusual or bizarre that makes them realize they
are dreaming. Participants were told to repeat these steps until they either fell
asleep or their intention was set. If their mind wandered, they were told to
repeat the procedure so that the last thing they thought about was their
intention to remember to recognize the next time they are dreaming. Partici-
pants were told that the longer they spend doing the technique the more
effective it would be, and not to worry if it took a long time to fall asleep.
Procedure
Participants accessed the online pretest questionnaire using a web URL
included in all promotional materials and media items. The questionnaire was
hosted by the survey management website Survey Monkey and was configured
so that participants could not navigate back to change their answers. Partici-
pants provided postal details so they could be sent materials via post.
Participants thus completed the study in their own homes. Participants were
randomly allocated to the nine possible combinations of the three Week 1 and
three Week 2 groups. There was no significant difference between the number
of participants in these nine combinations: ␹
2
(4, N⫽169) ⫽1.89, p⫽.756.
Participants were told that the purpose of Week 1 was to gather baseline
information about normal sleeping patterns and dream recall ability, and were
asked not to attempt any lucid dream induction techniques or to improve their
dream recall during this period. Participants filled in their logbooks immediately
upon waking and were urged to complete all seven logbook days consecutively
if possible. However, during Week 2 they were told that it is better to skip a day
if they were feeling sleep deprived and to make up for it at the end. During
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 215

Week 2, participants practiced lucid dream induction techniques as per the
instructions described in Section Lucid dream induction technique documents.
Participants returned their completed logbooks using prepaid envelopes pro-
vided.
Results
Preliminary Analyses and Descriptive Statistics
Most variables were not normally distributed and nonparametric tests were
used in all cases. Mann–Whitney tests indicated that the three Week 2 lucid
dreaming groups did not differ on any of the pretest or Week 1 variables (for the
sake of brevity, these analyses are not reported here). The ratio of males to females
did not differ between participants who did and did not complete the full study:
␹
2
(1, N⫽418) ⫽1.30, p⫽.254. The proportions of participants who were
employed nonstudents, students, and unemployed or retired did not differ among
participants who did and did not complete the full study: ␹
2
(2, N⫽420) ⫽4.30, p⫽
.117. Participants took 7.8 (SD ⫽1.8) days on average to complete an average of 6.8
(SD ⫽0.9) logbook entries during Week 2. Mann–Whitney tests indicated that
participants who went on to complete the full study were not significantly different
from those who did not on any pretest variables except for being 6.4 years older on
average. In Week 2 of the study, participants spent (slightly) more time asleep
compared with Week 1. These findings are presented with descriptive statistics in
Table 1.
Independent samples Kruskal-Wallis tests were conducted to investigate group
differences in Week 2 logbook variables and are presented with descriptive
statistics in Table 2. Group differences reached statistical significance for two
variables: L time asleep, and L tiredness on waking. However, upon applying a
Bonferroni correction to control the Type I error rate and using an alpha level of
.006 (.05/9), none of the group comparisons were statistically significant. Thus, post
hoc comparisons were not conducted.
Relationships With Overall Lucid Dreaming Rates
Spearman rho nonparametric correlations were calculated to investigate
relationships between both pretest and mean Week 2 lucid dreaming rates and
other pretest and Week 2 variables, and are presented in Table 3.Itwas
hypothesized that there would be significant positive correlations between general
dream recall rates and lucid dreaming rates at both pretest and during Week 2. This
hypothesis was supported. All pretest dream recall variables were related to P lucid
DC (per month). Correlations between these pretest variables and L DRF lucid
were weaker and in most cases nonsignificant. This pattern was reversed for Week
2 general dream recall variables, which were more strongly and consistently
correlated with L DRF lucid than with P lucid DC (per month). These findings
highlight the importance of avoiding comparisons between retrospective and
logbook measures of dream recall (see Aspy, 2016). P lucid tech prac was weakly
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
216 ASPY, DELFABBRO, PROEVE, AND MOHR

Table 1
Descriptive Statistics for Pretest, Week 1, and Week 2 Variables With Mann–Whitney Tests for Pretest Differences Between Participants Who Did and Did Not Complete
the Full Study and Wilcoxon Signed-Ranks Tests Between Week 1 and Week 2 Logbook Variables
M(SD)
Mann–
Whitney test M(SD)
Wilcoxon
test
Pretest variable
Completed full
study
(N⫽169)
Did not com-
plete full study
(N⫽251) Zp Logbook variable
Week 1
(N⫽169)
Week 2
(N⫽169) Zp
P DRF 44.3% (30.0%) 40.0% (27.1%) 1.17 .244 L DRF 77.1% (21.4%) 77.1% (22.9%) 0.11 .912
P DC (per day) 0.6 (0.6) 0.6 (0.6) 1.61 .108 L DC (per day) 1.7 (1.1) 1.8 (1.2) 0.27 .791
P DC lucid (per month) 1.5 (4.0) 1.4 (3.7) 0.34 .738 L time asleep 7.5 (0.8) 7.6 (0.8) 3.12 .002
P lucid tech prior 36.1% (48.2%) 30.7% (46.2%) 1.16 .247 L sleep quality 3.5 (0.5) 3.5 (0.5) 1.67 .095
P lucid tech freq 0.3 (0.9) 0.3 (1.0) 0.36 .719 L tiredness on waking 2.4 (0.7) 2.4 (0.8) 0.87 .386
P age 38.3 (15.0) 31.9 (13.0) 4.39 ⬍.001 L sleep dep yesterday 1.9 (0.6) 1.9 (0.7) 0.48 .634
Note.P⫽pretest variable; L ⫽logbook variable; DRF ⫽dream recall frequency; DC ⫽dream count.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 217

correlated with P lucid DC (per month) but not with Week 2 L DRF lucid,
indicating that prior experience with lucid dream induction techniques did not
influence the effectiveness of the techniques used in the present study. There was
a significant positive correlation between pretest and Week 2 lucid dreaming rates.
However, shared variance was only 9.6%. Age was positively correlated with both
pretest and Week 2 lucid dreaming rates.
Lucid Dream Induction
It was hypothesized that lucid dreaming rates would be significantly higher in
Week 2 compared with Week 1 for all participants combined and for participants
in each of the three Week 2 groups. This hypothesis was partially supported. As can
be seen in Table 4, dependent samples Wilcoxon’s tests showed that L DRF lucid
Table 2
Kruskal–Wallis Tests for Differences Between the RT Only, RT ⫹WBTB, and RT ⫹WBTB ⫹MILD
Groups in Week 2 Logbook Variables
M(SD)
Kruskal–
Wallis test
Logbook variable
All participants
(n⫽169)
RT only
(n⫽68)
RT ⫹WBTB
(n⫽54)
RT ⫹
WBTB ⫹
MILD
(n⫽47) ␹
2
p
L lucid duration minutes 11.3 (15.6) 9.2 (8.6) 13.0 (14.9) 11.6 (20.8) 1.96 .376
L reality tests 10.1 (4.5) 10.6 (5.6) 10.2 (3.7) 9.2 (3.5) 1.69 .429
L DRF 77.1% (22.9%) 75.7% (24.3%) 75.7% (25.7%) 81.4 (18.6) 0.41 .815
L DC (per day) 1.8 (1.2) 1.7 (1.2) 1.8 (1.2) 1.8 (1.1) 0.81 .668
L DQ 5.6 (5.7) 5.3 (5.8) 5.0 (4.4) 6.6 (6.7) 3.26 .196
L time asleep 7.6 (0.8) 7.6 (0.9) 7.9 (0.8) 7.4 (0.7) 7.33 .026
L sleep quality 3.5 (0.5) 3.5 (0.5) 3.4 (0.5) 3.4 (0.6) 3.68 .159
L tiredness on waking 2.4 (0.8) 2.2 (0.7) 2.5 (0.7) 2.5 (0.8) 9.08 .011
L sleep dep yesterday 1.9 (0.7) 1.8 (0.6) 1.9 (0.7) 1.9 (0.8) 0.90 .638
Note.L⫽logbook variable; RT ⫽reality testing; WBTB ⫽wake back to bed; MILD ⫽mnemonic
induction of lucid dreams; DRF ⫽dream recall frequency; DC ⫽dream count.
Table 3
Spearman Rho Nonparametric Correlations Between Pretest and Week 2 Lucid Dreaming Rates and
Other Pretest and Week 2 Variables
P DC lucid (per month) Week 2 DRF lucid
P DC lucid (per month) .31
**
P lucid tech freq .11
*
⫺.10
P age .10
*
.31
**
P DRF .28
**
.15
P DC (per day) .27
**
.21
**
L DRF ⫺.04 .08
*
L DC (per day) .07 .20
**
L DQ .21
**
.25
**
Note.P⫽pretest variable; L ⫽logbook variable; WBTB ⫽wake back to bed; MILD ⫽mnemonic
induction of lucid dreams; DRF ⫽dream recall frequency; DC ⫽dream count.
*
p⬍.05.
**
p⬍.01.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
218 ASPY, DELFABBRO, PROEVE, AND MOHR

was significantly higher in Week 2 than in Week 1 for all participants combined.
The same was true for participants in the RT ⫹WBTB ⫹MILD group. L DRF
lucid was higher in the RT ⫹WBTB group in Week 2 compared with Week 1, but
this difference was not statistically significant. L DRF lucid was slightly lower in
Week 2 for participants in the RT only group, indicating that reality testing on its
own was not effective at inducing lucid dreams. An independent samples Kruskal-
Wallis test indicated that there were significant group differences in Week 2 L DRF
lucid (␹
2
⫽6.35, p⫽.042). Post hoc pairwise comparisons revealed that the
difference between the RT only and RT ⫹WBTB ⫹MILD groups was significant
(␹
2
⫽21.10, p⫽.035). However, the differences between the RT only and RT ⫹
WBTB groups (␹
2
⫽8.84, p⫽.816) and the RT ⫹WBTB and RT ⫹WBTB ⫹
MILD groups (␹
2
⫽12.26, p⫽.492) were nonsignificant.
In addition to L DRF lucid, lucid dreaming was also operationalized as L lucid
participants. A McNemar’s test showed that for all participants combined, the
proportion of participants that experienced lucid dreaming at least once during
Week2(L lucid participants ⫽44.6%) was significantly higher than in Week 1 (L
lucid participants ⫽27.7%): ␹
2
(1, N⫽166) ⫽13.50, pⱕ.001. A binomial
test—used because cell counts did not permit McNemar’s test—showed that the
increase in L lucid participants was significant in the RT ⫹WBTB ⫹MILD group
(Week 1 ⫽27.7%, Week 2 ⫽53.2%, p⫽.012). The increases in L lucid participants
were smaller and did not reach statistical significance in the RT only (Week 1 ⫽
25.4%, Week 2 ⫽36.8%, p⫽.077) and the RT ⫹WBTB (Week 1 ⫽30.8%, Week
2⫽46.3%, p⫽.096) groups. These findings provide partial support for the
hypothesis that lucid dreaming rates would be significantly higher in Week 2
compared with Week 1 for all participants combined and for participants in each of
the three Week 2 groups. A 3 ⫻2 Chi
2
test was performed to explore group
differences in L lucid participants. Results indicated that there were no statistically
significant group differences, ␹
2
(2, N⫽169) ⫽3.16, p⫽.206.
Relationships With Technique Practice Variables
A Spearman rho nonparametric correlation using data from each individual
logbook day indicated that, for all participants combined, the number of reality
Table 4
Improvements in Lucid Dreaming Rates in Week 2 Compared With Week 1 for All Participants Combined
and for Participants in Each Week 2 Group
Week 2 group
L DRF Lucid
Wilcoxon
test
Week 1
M(SD)
Week 2
M(SD) Improvement Zp
All participants (n⫽169) 8.1% (17.8%) 11.3% (17.3%) 39.0% 2.27 .023
RT only (n⫽68) 8.1% (17.8%) 7.6% (13.0%) ⫺6.8% .27 .786
RT ⫹WBTB (n⫽54) 6.9% (15.7%) 10.7% (16.3%) 54.1% 1.04 .301
RT ⫹WBTB ⫹MILD (n⫽47) 9.4% (20.0%) 17.4% (22.0%) 84.5% 2.94 .003
Note.L⫽logbook variable; RT ⫽reality testing; WBTB ⫽wake back to bed; MILD ⫽mnemonic
induction of lucid dreams; DRF ⫽dream recall frequency.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 219

tests performed was not related to whether participants experienced lucid dreaming
(r
s
⫽.05, p⫽.078, N⫽1087). This was also the case for participants in the RT only
group (r
s
⫽.03, p⫽.501, n⫽445) and the RT ⫹WBTB ⫹MILD group (r
s
⫽⫺.02,
p⫽.792, n⫽281). However, a significant correlation was observed for participants
in the RT ⫹WBTB group (r
s
⫽.17, pⱕ.001, n⫽361). These findings are given
greater consideration in Section Group Differences in Lucid Dream Induction.
Spearman rho nonparametric correlations between L DRF lucid and variables that
operationalize the way in which the WBTB and MILD techniques were practiced
are presented with descriptive statistics in Table 5. In both the RT ⫹WBTB and
RT ⫹WBTB ⫹MILD groups,L DRF lucid was higher when participants had less
difficulty focusing. L DRF lucid was also positively correlated with motivation to
practice the technique, but only in the RT ⫹WBTB group. Participants fell asleep
while performing the MILD technique in the majority (79.9%) of cases. A 2 ⫻2
chi-square test indicated that this was not related to the likelihood of experiencing
lucid dreaming: ␹
2
(1, n⫽293) ⫽0.48, p⫽.487. However, a significant negative
correlation was observed between L min back to sleep and L DRF lucid (see Table
5). Indeed, this relationship is stronger than any of the other relationships with L
DRF lucid observed in the present study. To further explore this relationship,
occasions when participants did not fall asleep while performing the MILD
technique and then took five minutes or less to fall asleep afterward were examined.
This was achieved a total of 24 times by 14 participants. For these 24 occasions,
Week 2 L DRF lucid (M⫽45.8%, SD ⫽50.9%) was much higher than for all the
other nights on which these participants practiced MILD (M⫽24.6%, SD ⫽
43.4%), suggesting that the MILD technique is most effective when sleep is
achieved within five minutes of completing the technique. However, it should be
noted that the baseline L DRF lucid rate for these participants during Week 1 was
higher than average at M⫽20.4% (SD ⫽29.5%), which limits the generalizability
of these findings. Notwithstanding, completing the technique and then falling asleep
within five minutes was associated with an increase in L DRF lucid of 86.2%
compared with all other MILD attempts for these 14 participants. Week 2 L DRF
lucid was significantly lower when participants performed more MILD technique
repetitions and spent longer doing so. However, when participants who fell asleep
Table 5
Spearman Rho Nonparametric Correlations Between Week 2 Lucid Dreaming Rates and Variables That
Operationalize the Way in Which the WBTB and MILD Techniques Were Practiced
RT ⫹WBTB group RT ⫹WBTB ⫹MILD group
M(SD)
Correlation (r
s
)
with L DRF
Lucid M(SD)
Correlation (r
s
)
with L DRF
Lucid
L difficulty focusing 2.7 (1.0) ⫺.11
*
3.0 (0.9) ⫺.12
*
L technique motivation 2.8 (1.0) .11
*
3.0 (1.0) ⫺.03
L minutes back to sleep 18.6 (14.3) .11
*
24.0 (25.7) ⫺.44
**
L technique repetitions 17.4 (16.1) ⫺.12
*
L technique min 8.5 (5.3) ⫺.16
**
Note. All correlations are point-biserial and based on daily observations. L ⫽logbook variable; RT ⫽
reality testing; WBTB ⫽wake back to bed; MILD ⫽mnemonic induction of lucid dreams; DRF ⫽
dream recall frequency.
*
p⬍.05.
**
p⬍.01.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
220 ASPY, DELFABBRO, PROEVE, AND MOHR

while performing the technique were excluded, the correlations with L technique
repetitions (r
s
⫽⫺.09, p⫽.168, n⫽223) and L technique min (r
s
⫽⫺.13, p⫽.057,
N⫽234) became smaller and nonsignificant. This is given greater consideration in
Section Technique repetitions and time spent on the technique.
Additional Exploratory Analyses
An independent samples Kruskal-Wallis test indicated that the type of logbook
used in Week 1 was not related to Week 2 L DRF lucid:␹
2
(2, N⫽169) ⫽0.89, p⫽
.641. Similarly, a 3 ⫻2 Chi
2
test indicated that there was no significant difference
in the proportion of participants in each Week 1 logbook group that experienced
lucid dreaming during Week 2: ␹
2
(2, N⫽169) ⫽1.18, p⫽.554. Thus, writing out
one’s dreams for a week prior to practicing lucid dreaming techniques did not
appear to be advantageous.
As noted in Section Relationships With Overall Lucid Dreaming Rates, for all
participants combined P lucid tech freq was not significantly correlated with Week
2L DRF lucid. The same was true for participants in the RT ⫹WBTB ⫹MILD
group (r
s
⫽⫺.21, p⫽.159). A 2 ⫻2 Chi
2
test indicated that whether participants
experienced lucid dreaming during Week 2 was also not significantly related to
whether they had attempted to practice lucid dream induction techniques previ-
ously (L lucid participants ⫽39.3%) or not (L lucid participants ⫽47.2%): ␹
2
(1,
N⫽169) ⫽0.98, p⫽.322. The same was true for participants in the RT ⫹WBTB ⫹
MILD group (previous experience: L lucid participants ⫽50.0%; no previous
experience: L lucid participants ⫽54.3%): ␹
2
(1, N⫽47) ⫽0.07, p⫽.797. A
Mann–Whitney test indicated that Week 2 L DRF lucid did not differ between
participants with previous experience (M⫽9.4%, SD ⫽16.5%) and without
previous experience (M⫽12.4%, SD ⫽17.7%): Z(169) ⫽0.51, p⫽.692. The same
was true for participants in the RT ⫹WBTB ⫹MILD group (previous experience:
M⫽9.5%, SD ⫽11.2%; no previous experience: M⫽20.1%, SD ⫽24.2%): Z(1,
N⫽47) ⫽0.81, p⫽.480.
A2⫻2 Chi
2
test indicated that in the RT ⫹WBTB ⫹MILD group, lucid
dreaming was associated with whether or not participants reported that they were
dreaming when they were awakened by their alarm to perform the technique, ␹
2
(2,
n⫽285) ⫽7.16, ␸⫽.16, p⫽.028. Lucid dreaming was most likely when
participants reported that they were not dreaming (n⫽75, L DRF lucid ⫽25.3%),
was less likely when participants were awakened while dreaming (n⫽96, LDRF
lucid ⫽17.7%), and was least likely when participants were unsure of whether or
not they were dreaming (n⫽114, L DRF lucid ⫽10.5%). However, because of the
disproportionately large number of occasions when participants were unsure, it
remains unclear whether or not waking from a dream before performing lucid
dream induction techniques was related to lucid dreaming. This issue is given
greater consideration in Section Sleep stage awakening. The same pattern of
findings was observed for the RT ⫹WBTB group, but was not statistically
significant, ␹
2
(2, n⫽358) ⫽2.53, ␸⫽.08, p⫽.282.
To further investigate factors that influenced the success rate of the MILD
technique, differences in logbook variables between nights when MILD was and
was not followed by lucid dreaming were investigated. These are presented in
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 221

Table 6. No significant differences were found for L reality tests and L technique
motivation. When MILD was followed by lucid dreaming, participants had signifi-
cantly less difficulty focusing, took less time to get back to sleep once they finished
the technique, performed fewer technique repetitions, and spent slightly less time
on the technique overall. However, upon applying a Bonferroni correction to
control the Type I error rate and using an alpha level of .004 (.05/12), the findings
regarding L difficulty focusing and L technique repetitions became nonsignificant.
Findings regarding these two variables should thus be interpreted with caution.
When MILD was followed by lucid dreaming, participants were significantly less
sleep deprived the previous day, had marginally better sleep quality, and were
significantly less tired the next morning. After applying the Bonferroni correction,
only the finding regarding L tiredness on waking remained significant. These
findings indicate that successful induction of lucid dreaming using the MILD
technique did not compromise sleep quality. General dream recall was superior
according to both measures used.
Discussion
The purpose of the present study was to provide a thorough empirical
investigation into the effectiveness of three widely used cognitive lucid dream
induction techniques—reality testing, WBTB, and the MILD technique. A pretest
questionnaire was administered, and baseline logbook data were collected during
Week 1. In Week 2, participants were randomly allocated to conditions that
involved reality testing only (RT only), reality testing and WBTB (RT ⫹WBTB),
and reality testing, WBTB and the MILD technique (RT ⫹WBTB ⫹MILD).
Results showed that the RT ⫹WBTB ⫹MILD condition was the most effective at
Table 6
Mann–Whitney Tests for Differences in Week 2 Logbook Variables Between Nights When Lucid
Dreaming Did and Did Not Occur Following Practice of the MILD Technique
Logbook variable
M(SD)
Mann–Whitney
test
Lucid dreaming
reported
(n⫽52)
No lucid dreaming
reported
(n⫽251) Zp
L reality tests 9.3 (5.8) 9.5 (3.8) 0.26 .792
L difficulty focusing 2.6 (1.3) 3.0 (1.2) 2.08 .038
L technique motivation 3.1 (1.3) 3.0 (1.4) 0.57 .572
L minutes back to sleep 10.6 (21.1) 26.7 (27.4) 3.34 .001
L technique repetitions 13.9 (16.9) 18.7 (22.1) 2.05 .040
L technique minutes 7.7 (10.2) 8.8 (7.6) 2.76 .006
L DC (per day) 2.9 (2.1) 1.7 (1.4) 3.74 ⬍.001
L DQ 14.4 (15.9) 5.3 (5.7) 4.92 ⬍.001
L time asleep 7.6 (1.4) 7.4 (1.1) 1.13 .260
L sleep quality 3.7 (0.8) 3.4 (0.9) 1.96 .050
L tiredness on waking 2.0 (0.9) 2.6 (1.1) 3.77 ⬍.001
L sleep dep yesterday 1.6 (0.8) 2.0 (1.1) 2.73 .006
Note.L⫽logbook variable; DC ⫽dream count; DQ ⫽dream quantity; MILD ⫽mnemonic induction
of lucid dreams.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
222 ASPY, DELFABBRO, PROEVE, AND MOHR

inducing lucid dreams. Results showed that the effectiveness of the MILD
technique was influenced by several factors, including general dream recall and the
amount of time taken to fall asleep after finishing the technique.
Group Differences in Lucid Dream Induction
The RT ⫹WBTB ⫹MILD condition was clearly the most effective at inducing
lucid dreams. Participants in this group had a mean L DRF lucid of 17.4%, which
is 84.5% higher than in Week 1. Furthermore, just over half of these participants
(53.2%) experienced lucid dreaming at least once during Week 2, which is nearly
twice as many as in Week 1. In the RT ⫹WBTB group, mean L DRF lucid in Week
2 was 10.7%, which is 54.1% higher than in Week 1. However, this increase was less
than that observed in the RT ⫹WBTB ⫹MILD group and was not statistically
significant. The purpose of the RT ⫹WBTB group was to control for the effects of
waking up after five hours of sleep, thinking about lucid dreaming, and expecting
that one might have a lucid dream as a consequence of practicing a lucid dream
induction technique. Findings indicate that these effects are at most only partly
responsible for the effectiveness of the MILD technique and thus support the
theory that the MILD technique works by creating a mnemonic intention to
remember that one is dreaming that is then later recalled during a nonlucid dream.
The present study replicates several earlier field studies that have shown the MILD
technique to be effective for inducing lucid dreams (Edelstein & LaBerge, 1992;
LaBerge, 1988;LaBerge et al., 1994;Levitan, 1989,1990a,1990b,1991;Levitan &
LaBerge, 1994;Levitan et al., 1992). With the exception of one laboratory study
(Kueny, 1985), the present study is the first investigation of the MILD technique
that was not conducted by the creator of the MILD technique LaBerge or any of
his research associates. Because it was clearly the most effective, the MILD
technique is given in-depth consideration in Section Findings Related to the MILD
Technique.
Contrary to hypotheses, reality testing on its own was not effective at inducing
lucid dreams. Mean L DRF lucid in the RT only group was only 7.6% in Week 2,
which was lower than the rate observed in Week 1. The percentage of participants
that experienced lucid dreaming at least once during Week 2 was the lowest in this
group at 36.8%, and was not significantly higher than in Week 1. These findings are
consistent with those of Taitz (2011) and LaBerge (1988), who also found that
reality testing on its own was ineffective. However, the present findings are at odds
with other studies that have found reality testing to be effective (Levitan, 1989;
Purcell, 1988;Purcell et al., 1986;Schlag-Gies, 1992). One possible explanation is
that it takes longer than a single week (as in the present study) for the practice of
reality testing to become habitual enough that performing reality tests while
dreaming becomes likely. Reality testing was found to be effective when practiced
for three weeks in the studies by Purcell (1988) and Purcell et al. (1986), and for
eight weeks in the study by Schlag-Gies (1992).
Another possible explanation for the diversity in findings is that reality testing
is substantially more effective if combined with other activities that assist with lucid
dream induction. For example, Purcell (1988) observed significantly higher lucid
dreaming rates in an experimental condition that involved reality testing, reading
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 223

over written dream narratives and becoming familiar with recurring anomalies that
may serve as triggers for lucidity. Similar results were reported by Paulsson and
Parker (2006), who asked participants to form the intention to have a lucid dream
directly before going to sleep in addition to practicing reality testing throughout the
day. This explanation is also supported by findings from the present study. The
correlation between L DRF lucid and the number of reality tests performed was
nonsignificant in the RT only group. However, the correlation was highly significant
in the RT ⫹WBTB group, which involved reality testing as well as waking up after
five hours of sleep, reading about lucid dreaming and then performing a reality test
before returning to sleep (see Section RT WBTB group). This may have primed
participants to perform reality tests shortly before REM sleep. In the RT ⫹
WBTB ⫹MILD group, this priming effect may have been negated by practicing the
MILD technique. This might explain why the number of reality tests performed was
not correlated with L DRF lucid in this group.
Findings Related to the MILD Technique
General dream recall. General dream recall was correlated with L DRF lucid
and was significantly higher on occasions when practicing the MILD technique was
followed by lucid dreaming. There are two likely explanations for this. The first is
that heightened dreaming activity that leads to greater general dream recall is
conducive to lucid dream induction. The other is that the occurrence of lucid
dreams simply inflates general dream recall rates due to them being more vivid and
memorable than most nonlucid dreams. Participants recalled 1.2 more dreams on
average when practicing the MILD technique led to lucid dreaming, which at first
glance appears consistent with the explanation that general dream recall rates were
simply inflated by the occurrence of highly memorable lucid dreams. However,
most lucid dreams occur during the course of nonlucid dreams and it is likely that
at least a substantial portion of these dreams would have been recalled even if
lucidity had not been attained. In light of this, the finding that participants recalled
1.2 more dreams on average when they experienced lucid dreaming tentatively
suggests that there is indeed an effect whereby heightened dreaming activity that
gives rise to superior general dream recall is conducive to lucid dream induction. If
this is correct, it may be possible to increase the effectiveness of cognitive lucid
dream induction techniques such as the MILD technique by enhancing dreaming
activity during the night. This is given further consideration in Section Directions
for Future Research.
Sleep stage awakening. Lucid dreaming was most likely when participants
reported that they were not dreaming when they were awakened by their alarm to
perform the MILD technique. Lucid dreaming was less likely if participants
reported that they were dreaming and least likely when they were unsure of
whether or not they had been dreaming. However, it remains uncertain whether
waking up during a dream influences the effectiveness of the MILD technique. This
is because participants were unsure of whether they had been dreaming in the
majority of cases. If participants were not dreaming on most of the occasions when
they were unsure, this would mean that lucid dreaming was less likely following
awakening from dreamless sleep than the results indicate. On the other hand, the
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
224 ASPY, DELFABBRO, PROEVE, AND MOHR

fact that self-reported sleep stage awakening was significantly related to LDRF
lucid in the RT ⫹WBTB ⫹MILD group but not the RT ⫹WBTB group suggests
that sleep stage awakening may indeed influence the effectiveness of the MILD
technique in some way. This is given further consideration in Section Directions for
Future Research.
Technique repetitions and time spent on the technique. On occasions when
performing the MILD technique was followed by lucid dreaming, participants
performed fewer technique repetitions and spent less time on the technique.
Furthermore, significant negative correlations were observed between LDRF
lucid and both the number of technique repetitions and the amount of time spent
on the technique. These findings appear counterintuitive, as one would expect that
more technique repetitions and more time spent on the technique would assist in
creating a strong mnemonic intention to remember that one is dreaming. However,
upon closer inspection it was found that difficulty falling asleep after completing the
technique was strongly related to both the number of technique repetitions (r
s
⫽
.48, pⱕ.001) and the amount of time spent on the technique (r
s
⫽.67, pⱕ.001).
Furthermore, the correlations between L DRF lucid and both technique repetitions
and time spent on the technique became smaller and nonsignificant when they were
recalculated using only occasions when participants fell asleep before completing
the technique. These findings are consistent with the theory that the effectiveness
of the MILD technique is highly dependent on being able to fall asleep quickly after
creating a strong mnemonic intention to remember that one is dreaming. It appears
that this mnemonic intention tends to become weaker when one takes longer to fall
asleep after completing the technique.
Time taken to return to sleep. The strongest predictor of lucid dreaming
following practice of the MILD technique was the amount of time it took for
participants to fall asleep after they finished the technique. On occasions when
participants were able to fall asleep in less than five minutes, L DRF lucid was very
high at M⫽45.8%. This rate is 86.2% higher than the mean L DRF lucid rate for
all other nights on which these participants attempted the MILD technique (M⫽
24.6%). It should be noted that the Week 1 base rate for the 14 participants who
managed to achieve this (M⫽20.4%) was substantially higher than for all
participants in the RT ⫹WBTB ⫹MILD group (M⫽9.4%), which calls into
question the generalizability of this finding. Notwithstanding, if this amount of
improvement (86.2%) were extrapolated to all participants in the RT ⫹WBTB ⫹
MILD group, this would yield a mean L DRF lucid rate of 32.4%. This is an
exciting possibility because a lucid dream induction rate of this magnitude would
make research into the potential applications of lucid dreaming highly feasible. A
potential problem is that it may be difficult to fall asleep within five minutes of
completing the MILD technique. Indeed, participants fell asleep before finishing
the technique in the majority (79.9%) of cases. However, participants were told that
it did not matter if they fell asleep while practicing the MILD technique or how
long it took to fall asleep afterward. With altered instructions it may be possible to
reduce both the likelihood of falling asleep prematurely and the amount of time
required to fall asleep after completing the technique. This is given further
consideration in Section Directions for Future Research.
Relationships with sleep quality. On occasions when practicing the MILD
technique was followed by lucid dreaming, participants reported being significantly
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 225

less sleep deprived the previous day. This suggests that the MILD technique may
be more effective if one is well-rested. However, this finding became nonsignificant
after applying a Bonferroni correction, and must be interpreted with caution. On
occasions when MILD led to lucid dreaming, participants reported that they felt
significantly less tired upon waking. There were no other significant differences in
sleep quality variables. Therefore, it appears that successful lucid dream induction
using the MILD technique was not detrimental to sleep quality.
Accessibility of the MILD technique. Participants with prior lucid dreaming
experience were no more likely to experience lucid dreaming and did not have
higher mean L DRF lucid during Week 2 compared with participants with no prior
experience, nor was L DRF lucid correlated with the frequency of previous lucid
dream induction technique practice. This was the case for all participants combined
and for participants in the RT ⫹WBTB ⫹MILD group specifically. Pretest lucid
dreaming rates were significantly correlated with Week 2 L DRF lucid, but shared
variance was only 9.6%. Shared variance was similarly low for participants in the
RT ⫹WBTB ⫹MILD group at 11.2% (r
s
⫽.33, p⫽.022). These findings indicate
that it is not necessary to have prior experience with lucid dream induction
techniques in order for the MILD technique to be effective, nor is it necessary to
be a naturally prolific lucid dreamer. Thus, the MILD technique appears to be
appropriate for people who are naive to lucid dreaming and effective within a short
period of time.
Strengths and Limitations
The present study is the most methodologically rigorous lucid dream induction
field study ever conducted and is based on a relatively large and highly diverse
sample of participants from across Australia. The majority of previous lucid dream
induction studies have used participants that were either self-selected lucid
dreamers or undergraduate students. In contrast, the majority (63.9%) of partici-
pants in the present study had never attempted a lucid dream induction technique
before and only 21.3% of participants were students. Although most participants
who completed the pretest questionnaire did not go on to complete the full study,
those who did were comparable to those who did not on all pretest variables, except
for being significantly older (by 6.4 years on average). Thus, it appears that findings
from the present study are generalizable to a wide range of people that are
interested in learning to have lucid dreams. The present study has high ecological
validity because participants trialed the techniques in their own homes using
written instructions and without any contact from the experimenters, which reflects
how people typically learn lucid dream induction techniques. A limitation of the
present study is that the MILD technique was not trialed in isolation from reality
testing. This was done in the interests of identifying a maximally effective approach
to lucid dream induction. Although reality testing on its own was found to be
ineffective in the RT only group, the possibility that reality testing contributed to
the L DRF lucid rate observed in the RT ⫹WBTB ⫹MILD group cannot be ruled
out. Further research comparing MILD on its own to MILD combined with reality
testing would shed light on this. Another limitation is that lucid dreaming
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
226 ASPY, DELFABBRO, PROEVE, AND MOHR

techniques were only practiced for one week. This may not have been enough time
for reality testing to become effective.
Directions for Future Research
Findings from the present study indicate that the effectiveness of the MILD
technique could be improved with strategies designed to help participants develop
a strong mnemonic intention to remember that they are dreaming and then fall
asleep quickly without losing this intention. A one-size-fits-all approach is not likely
to be effective and participants will probably need to be given a range of strategies
for achieving an ideal level of wakefulness. For participants who are prone to falling
asleep prematurely, turning on lights, spending more time reading about lucid
dreaming, getting out of bed for a short period of time, or even writing out the
phrase “next time I’m dreaming, I will remember that I’m dreaming” multiple times
on paper may be helpful. In contrast, for participants who find it difficult to fall
asleep after completing the technique, it will be important to minimize such
stimulation. It remains unclear whether sleep stage awakening influences the
effectiveness of the MILD technique. This issue is worth further investigation and
this could be done in a sleep laboratory by comparing the effectiveness of the
MILD technique following awakenings from various sleep stages. If there is indeed
a sleep stage awakening that is most conducive to lucid dream induction, practi-
tioners of lucid dream induction techniques could take advantage of this knowledge
in the home setting using recently developed software applications that track sleep
activity using the accelerometers in smartphones. Although these software appli-
cations are less accurate than the equipment used in sleep laboratories, they permit
users to set alarms that go off when a specific sleep stage is detected. Users could
set an alarm to wake them up during a sleep stage that is most conducive to lucid
dreaming, thus increasing the effectiveness of cognitive lucid dream induction
techniques that involve an awakening such as the MILD technique.
In the present study, general dream recall was higher when practicing the
MILD technique was followed by lucid dreaming and was also correlated with
overall L DRF lucid rates. This indicates that heightened dreaming activity is
conducive to lucid dreaming. There is an abundance of anecdotal reports on the
extensive online lucid dreaming forums (e.g., Dream Views,LD4all,World of Lucid
Dreaming) indicating that certain substances are highly effective for increasing
dreaming activity and also for inducing lucid dreams (see also Yuschak, 2006).
Some of these substances influence the REM-on neurotransmitter acetylcholine
and include acetylcholine esterase inhibitors such as Galantamine, Huperzine-A
and Donepezil. To date, three studies have investigated the use of such substances
for inducing lucid dreams. In a pilot study by LaBerge (2004), it was found that the
odds ratio of experiencing lucid dreaming was extremely high at 0.75 on nights
when 10 mg of Donepezil was administered, compared with only 0.03 for
participants in a placebo condition. However, adverse effects including insomnia,
sleep paralysis, and gastrointestinal symptoms were reported in some cases. In an
unpublished study by LeMarca and LaBerge (2012, as cited in Sparrow, Hurd, &
Carlson, 2016), participants who ingested Galantamine during a brief awakening
(dose not specified) purportedly experienced a fivefold increase in lucid dreaming
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 227

compared with participants in a placebo condition. Most recently, participants in a
survey of 19 lucid dreaming enthusiasts who used Galantamine for inducing lucid
dreams reported that their Galantamine-induced lucid dreams were significantly
longer and more vivid than their other lucid dreams and contained significantly less
fear, threatening dream characters, violence, and darkness, with no increase in
sleep paralysis (Sparrow et al., 2016). Another potential dream-enhancing sub-
stance with less risk of adverse side effects is vitamin B6. Ebben, Lequerica, and
Spielman (2002) found that 240 mg of vitamin B6 (pyridoxine hydrochloride)
increased the vividness, emotionality, bizarreness, and color of dreams when
consumed before bed. However, this was only a small pilot study, and effects on
lucid dreaming were not reported. It may be possible to combine acetylcholine
esterase inhibitors, vitamin B6, or other potential dream-enhancing substances (see
Yuschak, 2006) with cognitive lucid dream induction techniques such as the MILD
technique to great effect, and research into this is certainly warranted.
External stimulation techniques represent another promising approach for
increasing the effectiveness of cognitive techniques. Light stimulation appears to be
the most effective (Stumbrys et al., 2012) and has been used in combination with
the MILD technique in four studies. Findings from these studies indicate that this
combination is more effective than the MILD technique on its own (LaBerge, 1988;
LaBerge & Levitan, 1995;LaBerge, Levitan, Rich, & Dement, 1988;Levitan &
LaBerge, 1994). Several commercially available devices designed to induce lucid
dreams in this way have been created by LaBerge’s research group, such as the
DreamLight,DreamLink, and NovaDreamer, and various generic versions exist.
These devices are designed to be used in the home setting and include an
eye-mask with sensors that detect the eye movements that characterize REM
sleep. Once REM sleep is detected, the device produces a series of flashing LED
lights within the mask that are incorporated into the dream experience and
serve as a cue that one is dreaming. With further research, the MILD technique
in conjunction with light stimulation and the administration of a dream-
enhancing substance could prove to be a highly effective approach to lucid
dream induction.
Conclusions
The present study indicates that the MILD technique is effective for
inducing lucid dreams within a short period of time and is suitable for a wide
range of people, including people that are naive to lucid dreaming. Based on the
present findings, several strategies for improving the effectiveness of the MILD
technique were identified. Combining the MILD technique with substances that
may enhance dreaming activity and with external stimulation may further
enhance the effectiveness of the MILD technique. Lucid dreaming has a wide
range of potential benefits and applications, and the only impediment to
research in this area is the lack of effective and reliable lucid dream induction
techniques. Thus, high quality empirical research on lucid dream induction
should be considered a high priority among dream researchers.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
228 ASPY, DELFABBRO, PROEVE, AND MOHR

References
Aspy, D. J. (2016). Is dream recall underestimated by retrospective measures and enhanced by keeping a
logbook? An empirical investigation. Consciousness and Cognition, 42, 181–203. http://dx.doi.org/10
.1016/j.concog.2016.03.015
Aspy, D. J., Delfabbro, P., & Proeve, M. (2015). Is dream recall underestimated by retrospective
measures and enhanced by keeping a logbook? A review. Consciousness and Cognition, 33,
364–374. http://dx.doi.org/10.1016/j.concog.2015.02.005
Brown, R., & Donderi, D. C. (1986). Dream content and self-reported well-being among recurrent
dreamers, past recurrent dreamers, and nonrecurrent dreamers. Journal of Personality and Social
Psychology, 50, 612–623. http://dx.doi.org/10.1037/0022-3514.50.3.612
Cheyne, J. A. (2003). Sleep paralysis and the structure of waking-nightmare hallucinations. Dreaming,
13, 163–179. http://dx.doi.org/10.1023/A:1025373412722
Dane, J. R. (1984). A comparison of waking instructions and posthypnotic suggestion for lucid dream
induction (Unpublished doctoral dissertation). Georgia State University, Atlanta, Georgia.
Downs, S. H., & Black, N. (1998). The feasibility of creating a checklist for the assessment of the
methodological quality both of randomised and non-randomised studies of health care interven-
tions. Journal of Epidemiology and Community Health, 52, 377–384. http://dx.doi.org/10.1136/jech
.52.6.377
Ebben, M., Lequerica, A., & Spielman, A. (2002). Effects of pyridoxine on dreaming: A preliminary
study. Perceptual and Motor Skills, 94, 135–140. http://dx.doi.org/10.2466/pms.2002.94.1.135
Edelstein, J., & LaBerge, S. (1992). The best time for lucid dreaming: Naps, mishaps, and recaps.
NightLight, 4, 4–9.
Erlacher, D., Schädlich, M., Stumbrys, T., & Schredl, M. (2014). Time for actions in lucid dreams: Effects
of task modality, length, and complexity. Frontiers in Psychology, 4, 1013. http://dx.doi.org/10.3389/fpsyg
.2013.01013
Erlacher, D., & Schredl, M. (2010). Practicing a motor task in a lucid dream enhances subsequent
performance: A pilot study. The Sport Psychologist, 24, 157–167. http://dx.doi.org/10.1123/tsp
.24.2.157
Fenwick, P., Schatzman, M., Worsley, A., Adams, J., Stone, S., & Baker, A. (1984). Lucid dreaming:
Correspondence between dreamed and actual events in one subject during REM sleep. Biological
Psychology, 18, 243–252. http://dx.doi.org/10.1016/0301-0511(84)90056-5
Fingerlin, T. (2013). Incidence and frequency of lucid dreams in a Swiss junior college student sample.
International Journal of Dream Research, 6, 127–129.
Gackenbach, J. (1991). An estimate of lucid dreaming incidence. Lucidity Letter, 10, 232–233.
Hearne, K. M. T. (1978). Lucid dreams: An electro-physiological and psychological study (Unpublished
doctoral dissertation). University of Liverpool, UK.
Hobson, A. (2009). The neurobiology of consciousness: Lucid dreaming wakes up. International Journal
of Dream Research, 2, 41–44.
Holzinger, B., Klösch, G., & Saletu, B. (2015). Studies with lucid dreaming as add-on therapy to Gestalt
therapy. Acta Neurologica Scandinavica, 131, 355–363. http://dx.doi.org/10.1111/ane.12362
Kueny, S. R. (1985). An examination of auditory cueing in REM sleep for the induction of lucid dreams
(Unpublished doctoral dissertation). Pacific Graduate School of Psychology, Palo Alto, California.
LaBerge, S. (1980). Lucid dreaming: An exploratory study of consciousness during sleep (Unpublished
doctoral dissertation). Stanford University, Stanford, California.
LaBerge, S. (1985). Lucid dreaming: The power of being awake and aware in your dreams. Los Angeles,
CA: Tarcher.
LaBerge, S. (1988). Induction of lucid dreams including the use of the Dreamlight. Lucidity Letter, 7,
15–21.
LaBerge, S. (2004). Substances that enhance recall and lucidity during dreaming. United States patent
application publication no. US 2004/0266659 A1.
LaBerge, S., & DeGracia, D. J. (2000). Varieties of lucid dreaming experience. In R. G. Kunzendorf &
B. Wallace (Eds.), Individual differences in conscious experience (pp. 269–307). Amsterdam,
Netherlands: John Benjamins. http://dx.doi.org/10.1075/aicr.20.14lab
LaBerge, S., & Levitan, L. (1995). Validity established of DreamLight cues for eliciting lucid dreaming.
Dreaming, 5, 159–168. http://dx.doi.org/10.1037/h0094432
LaBerge, S., Levitan, L., Rich, R., & Dement, W. C. (1988). Induction of lucid dreaming by light
stimulation during REM sleep. Sleep Research, 17, 104.
LaBerge, S., Phillips, L., & Levitan, L. (1994). An hour of wakefulness before morning naps makes
lucidity more likely. NightLight, 6, 1–4.
LaBerge, S., & Rheingold, H. (1991). Exploring the world of lucid dreaming. Mississauga, Ontario,
Canada: Ballantine Books.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 229

Lancee, J., van den Bout, J., & Spoormaker, V. I. (2010). Expanding self-help imagery rehearsal therapy
for nightmares with sleep hygiene and lucid dreaming: A waiting-list controlled trial. International
Journal of Dream Research, 3, 111–120.
Levitan, L. (1989). A comparison of three methods of lucid dream induction. NightLight, 1, 9–12.
Levitan, L. (1990a). The best time for lucid dreaming. NightLight, 2, 9–11.
Levitan, L. (1990b). Is fifteen minutes enough? It’s too soon to tell....NightLight, 2, 4–14.
Levitan, L. (1991). Get up early, take a nap, be lucid. NightLight, 3, 1–9.
Levitan, L., & LaBerge, S. (1994). Of the MILD technique & dream recall, of minds & dream machines.
NightLight, 6, 9–12.
Levitan, L., LaBerge, S., & Dole, J. (1992). Morning naps are better than afternoon naps for lucid
dreaming. NightLight, 4, 4–10.
Love, D. (2013). Are you dreaming? Exploring lucid dreams a comprehensive guide. London, UK:
Enchanted Loom.
Malamud, J. R. (1979). The development of a training method for the cultivation of “Lucid” awareness
in fantasy, dreams, and waking life (Unpublished doctoral dissertation). New York University, New
York.
Nielsen, T. A. (2000). A review of mentation in REM and NREM sleep: “covert” REM sleep as a
possible reconciliation of two opposing models. Behavioral and Brain Sciences, 23, 851–866.
http://dx.doi.org/10.1017/S0140525X0000399X
Ogilvie, R. D., Hunt, H. T., Tyson, P. D., Lucescu, M. L., & Jeakins, D. B. (1982). Lucid dreaming and
alpha activity: A preliminary report. Perceptual and Motor Skills, 55, 795–808. http://dx.doi.org/10
.2466/pms.1982.55.3.795
Paul, F., Schadlich, M., & Erlacher, D. (2014). Lucid dream induction by visual and tactile stimulation:
An exploratory sleep laboratory study. International Journal of Dream Research, 7, 61–66.
Paulsson, T., & Parker, A. (2006). The effects of a two-week reflection-intention training program on
lucid dream recall. Dreaming, 16, 22–35. http://dx.doi.org/10.1037/1053-0797.16.1.22
Purcell, S. (1988). The education of attention to dreaming in high and low frequency dream recallers: The
effects on dream self-reflectiveness lucidity and control (Unpublished doctoral dissertation). Carleton
University, Ottawa, Ontario, Canada.
Purcell, S., Mullington, J., Moffitt, A., Hoffmann, R., & Pigeau, R. (1986). Dream self-reflectiveness as
a learned cognitive skill. Sleep, 9, 423–437. http://dx.doi.org/10.1093/sleep/9.3.423
Reed, H. (1973). Learning to remember dreams. Journal of Humanistic Psychology, 13, 33–48.
http://dx.doi.org/10.1177/002216787301300305
Reis, J. (1989). Entwicklung Einer Biofeedback-Technik zur Induktion von Klartraumen [Development
of a biofeedback technique for the induction of lucid dreams]. Bewusst Sein, 1, 57–66.
Saunders, D. T., Roe, C. A., Smith, G., & Clegg, H. (2016). Lucid dreaming incidence: A quality effects
meta-analysis of 50 years of research. Consciousness and Cognition, 43, 197–215. http://dx.doi.org/10
.1016/j.concog.2016.06.002
Schlag-Gies, C. (1992). Untersuchung der Effektivität zur Induktion von Klarträumen [Investigation of
effectiveness for the induction of lucid dreams] (Unpublished doctoral thesis). Saarland University,
Germany.
Schredl, M. (2004). Reliability and stability of a dream recall frequency scale. Perceptual and Motor
Skills, 98, 1422–1426. http://dx.doi.org/10.2466/pms.98.3c.1422-1426
Schredl, M., & Erlacher, D. (2004). Lucid dreaming frequency and personality. Personality and
Individual Differences, 37, 1463–1473. http://dx.doi.org/10.1016/j.paid.2004.02.003
Schredl, M., & Erlacher, D. (2011a). Dream consciousness and sleep physiology. In D. Cvetkovic & I.
Cosic (Eds.), States of consciousness: Experimental insights into meditation, waking, sleep and
dreams (pp. 93–108). New York, NY: Springer-Verlag Publishing. http://dx.doi.org/10.1007/978-3-
642-18047-7_4
Schredl, M., & Erlacher, D. (2011b). Frequency of lucid dreaming in a representative German sample.
Perceptual and Motor Skills, 112, 104–108. http://dx.doi.org/10.2466/09.PMS.112.1.104-108
Sharpless, B. A., & Barber, J. P. (2011). Lifetime prevalence rates of sleep paralysis: A systematic
review. Sleep Medicine Reviews, 15, 311–315. http://dx.doi.org/10.1016/j.smrv.2011.01.007
Snyder, T. J., & Gackenbach, J. (1988). Individual differences associated with lucid dreaming. In J.
Gackenbach & S. LaBerge (Eds.), Conscious mind, sleeping brain: Perspectives on lucid dreaming
(pp. 221–259). New York, NY: Plenum Press. http://dx.doi.org/10.1007/978-1-4757-0423-5_10
Sparrow, G. S., Hurd, R., & Carlson, R. (2016). Assessing the perceived differences in post-Galantamine
lucid dreams vs. non-Galantamine lucid dreams. International Journal of Dream Research, 9.
Spoormaker, V. I., & van den Bout, J. (2006). Lucid dreaming treatment for nightmares: A pilot study.
Psychotherapy and Psychosomatics, 75, 389–394. http://dx.doi.org/10.1159/000095446
Stumbrys, T., & Daniels, M. (2010). An exploratory study of creative problem solving in lucid dreams:
Preliminary findings and methodological considerations. International Journal of Dream Research,
3, 121–129.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
230 ASPY, DELFABBRO, PROEVE, AND MOHR

Stumbrys, T., & Erlacher, D. (2012). Lucid dreaming during NREM sleep: Two case reports.
International Journal of Dream Research, 5, 151–155.
Stumbrys, T., Erlacher, D., Schädlich, M., & Schredl, M. (2012). Induction of lucid dreams: A systematic
review of evidence. Consciousness and Cognition, 21, 1456–1475. http://dx.doi.org/10.1016/j.concog
.2012.07.003
Stumbrys, T., Erlacher, D., & Schredl, M. (2013a). Reliability and stability of lucid dream and nightmare
frequency scales. International Journal of Dream Research, 6, 123–126.
Stumbrys, T., Erlacher, D., & Schredl, M. (2013b). Testing the involvement of the prefrontal cortex in
lucid dreaming: A tDCS study. Consciousness and Cognition, 22, 1214–1222. http://dx.doi.org/10
.1016/j.concog.2013.08.005
Stumbrys, T., Erlacher, D., & Schredl, M. (2016). Effectiveness of motor practice in lucid dreams: A
comparison with physical and mental practice. Journal of Sports Sciences, 34, 27–34. http://dx.doi.org/10
.1080/02640414.2015.1030342
Taitz, I. (2011). Learning lucid dreaming and its effect on depression in undergraduates. International
Journal of Dream Research, 4, 117–126.
Tholey, P. (1983). Techniques for inducing and manipulating lucid dreams. Perceptual and Motor Skills,
57, 79–90. http://dx.doi.org/10.2466/pms.1983.57.1.79
Yuschak, T. (2006). Advanced lucid dreaming: The power of supplements. Raleigh, NC: Lulu Enter-
prises.
E-Mail Notification of Your Latest Issue Online!
Would you like to know when the next issue of your favorite APA journal
will be available online? This service is now available to you. Sign up at
https://my.apa.org/portal/alerts/ and you will be notified by e-mail when issues of
interest to you become available!
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
REALITY TESTING AND THE MNEMONIC INDUCTION 231