Lucid dreaming: an age-dependent brain dissociation

Authors


  • This study was not supported by external funding. There is no conflict of interest of any author. The study is not used by any off-label or investigation.

Ursula Voss, PhD, Bonn University, Department of Psychology, Kaiser-Karl-Ring 9, 53111 Bonn, Germany. Tel.: +49 6175-791-559; fax: +49 69-53087-468; e-mail: u.voss@uni-bonn.de

Summary

The current study focused on the distribution of lucid dreams in school children and young adults. The survey was conducted on a large sample of students aged 6–19 years. Questions distinguished between past and current experience with lucid dreams. Results suggest that lucid dreaming is quite pronounced in young children, its incidence rate drops at about age 16 years. Increased lucidity was found in those attending higher level compared with lower level schools. Taking methodological issues into account, we feel confident to propose a link between the natural occurrence of lucid dreaming and brain maturation.

Introduction

The most obvious criterion for dream lucidity refers to the insight that the ongoing dream is only a dream and not reality. Sometimes, the dreamer can exert some control over the dream plot. In our earlier studies, we were able to identify some neurophysiological correlates of lucid dreaming in which part of the brain remains asleep while other parts regain almost waking-level function (Hobson and Voss, 2010, 2011; Voss et al., 2009). This neurophysiological dissociation is met on a subjective level with experiences of psychological dissociation, such as watching the dream as if from the outside or realizing that one is in a dream while the dream continues (Voss et al., submitted). In order to move from a mere correlational level to a better and more thorough understanding of lucid dreaming, we want to increase our efforts to learn more about its determinants. What exactly are its defining properties? At what ages does it set in? How and why does it occur naturally?

A limiting factor to the rigorous scientific study of lucid dreaming so far is the fact that its occurrence in adulthood is rather rare and difficult to maintain. This low incidence stands in apparent contrast to the high prevalence of reports claiming that the majority of adults have experienced lucidity in dreams at least once in their lifetime (Hearne, 1983; Schredl and Erlacher, 2011; Stepansky et al., 1998). Is it possible that adults report merely past experience but not current lucid dreaming status? Several authors have reported an inverse relationship of age and frequency of lucid dreaming (Blackmore, 1984; Schredl and Erlacher, 2004, 2011; Watson, 2001). Instead of interpreting this peculiarity as individual differences, as these authors have done, we pursued an alternative hypothesis, namely that lucid dreaming occurs primarily in childhood and puberty (Armstrong-Hickey, 1988; Hobson, 2009; Lapina et al., 1998).

In the following, we present preliminary evidence that the dissociative state of lucid dreaming is, indeed, at least coincidental with brain maturation during childhood and early adolescence. Specifically, we tested the hypothesis that lucid dreaming is less frequent in adults compared with children.

Materials and Methods

Participants

The original sample consisted of 793 students. We excluded 99 data sets (see Table 2), so that results are based on the answers of 694 students (346 female, 348 male) from schools in and around Bonn, Germany. Students attended primary schools (Grundschule, grades 1–4), as well as three types of secondary schools (Hauptschule, Realschule, Gymnasium). In Germany, following grade 4, students attend different types of secondary schools according to assessment of intellectual skills during primary school years (Bos and Stubbe, 2008; Maaz et al., 2008). The sample was representative for secondary school type attendance in Germany (Table 1).

Table 1.   Statistics for secondary school type attendance in the sample and in Germany
School levelDistribution in sample
n (%)
Distribution in Germany (%)Age range sample years
  1. Statistics of school type attendance in Germany were retrieved from the German Federal Census Department (2011).

Primary school75 (NA)NA6–10
Secondary school
 Lower level (Hauptschule)122 (22.3)18.19–17
 Medium level (Realschule)157 (28.7)28.49–18
 Higher level (Gymnasium)268 (49.0)53.59–19
Total622 (100.0)100.0 

Informed consent was taken from students or the parents of minors. Participants received flip image postcards or candy for their participation. In accordance with the regulations of the Ministry for Education in the state of Nordrheinwestfalen, the study was approved by the principals of all participating schools.

Procedure

One-on-one interviews were conducted by trained graduate students of the Psychology Department of Bonn University. Interviews were conducted in empty classrooms, during regular school hours. Students were collected conjointly from each classroom, one class at a time. While they awaited their turn to be interviewed, one of the interviewers tried to involve them in small-talk. After the interview, each student returned to class immediately.

Participating students were provided with a detailed description of lucid dreaming (lucid insight into the dream while remaining asleep) and ample time was given to make sure they had understood the concept. Students were then asked several sleep-related warm-up questions (Fig. 1), followed by seven dream-related ones. If a lucid dream was reported, the interviewer requested that an example of such a dream plot be shared (Narrative 1).

Figure 1.

 Interview questions regarding dream recall and frequency of lucid dreams.

Narrative 1 (boy, 7 years old): I dreamt I was playing soccer with my friends, and when I looked at my legs I saw that they were distorted. Then I realized it must be a dream because they did not at all look like my own legs. Then I looked up and saw that I was in a giant soccer stadium and I was able to play with my favorite soccer team (the adult team). I could run real fast, faster than in waking.

For validation purposes, the student was then asked to describe why he/she believed the dream had been a lucid one (Narrative 2).

Narrative 2 (boy, 11 years old): I knew I was dreaming because the houses and streets looked different from reality and I was able to climb up the walls, like Spiderman, just like that. And then I thought about what I could do up there on the roof, and because it was a dream, I jumped down from it.

Suggestibility and social compliance

Suggestibility was assessed with Questions 8 (Fig. 1) and 9 (Fig. 2). Affirmative answers to either question led to exclusion of the data set. In addition, we excluded data from students who gave contradictory answers to Questions 4 and 7 (Table 2).

Figure 2.

 (a) For Question 9, participants were shown the Lang stereocard (II) often used in ophthalmology for the assessment of strabismus. (b) Hidden objects: elephant, car, moon, star. After students described the objects, they were asked if they also could see the (non-existing) house on the top right. Note: this is a very crude test but it was quite effective with the young.

Table 2.   Descriptive statistics of the sample
Age (years)Participants, nSuggestible students
n (%)
Valid observations, nSex
FM
  1. Group sizes of students aged 8 and 9 years are smaller than those aged 10–15 years because we received fewer parental permissions to interview these children, most likely because they forgot to inform their parents. Students aged 16–19 years were more difficult to interview because of their variable schedules. Bold face indicates high values.

666 20 (30) 462026
739 13 (33) 261016
8355 (14)301317
9396 (15)331320
107810 (13)683731
119112 (13)794336
128310 (12)733241
131029 (10)933954
14707 (10)633528
15805 (6)754233
16341 (3)332112
172222148
18361 (3)351718
191818108
Total79399694346348

Regarding Questions 4 and 7, data of students who claimed to have last been lucid a week ago (Question 4) but affirmed Question 7 (Were you lucid last night?) were eliminated from further analysis.

Content validity of suggestibility questions is assumed because four independent experts agreed that the items possessed face validity in measuring this construct (Moosbrugger and Kelava, 2012). Item analysis further supports our validity assumption, showing that students who affirmed either of the two questions claimed lucidity more frequently than students who negated both questions (Question 7: T = 9.37, df = 791, P < 0.01; Question 8: T = 8.01, df = 791, P < 0.01). Whereas Question 7 was affirmed especially by young students aged 6–14 years, with a peak at ages 6 and 7 years, affirmative answers to Question 8 were equally distributed across ages 6–16 years.

We tried to control for social compliance by including only those interviews in which students either denied to have ever had a lucid dream or in which they provided a lucid dream narrative and a description of lucid elements that met the criterion of lucid insight (e.g. Narrative 2). Of course, especially older students may have heard or read about lucid dreaming and not given a valid report. However, we think it unlikely to have influenced younger students who reported the highest lucidity rates.

Data analysis

Univariate analysis of variance (anova) was conducted on interval scaled variables, using Sex and School Type as independent factors, and Age as covariate. Nominal scaled variables were analysed with t-tests or χ² procedures, depending on the scale of measurement of the dependent variable/s. Partial correlations were calculated to control for effects of Age. Dream recall frequency was analysed for the complete sample, while questions further specifying lucid dreams were based on lucid dreamers, only.

Results

The main findings of our survey were a surprisingly high incidence of reported lucidity in the young, and more frequent lucidity in those who are intellectually more capable. We also observed that the current incidence of lucid dreaming sharply decreased in early adulthood, while the experience of past lucid dreaming episodes increased steadily across age groups, suggesting that we must differentiate between the lifetime prevalence of a lucid dream and current lucid dreaming status.

Memory of lucid dreaming (entire sample)

In total, 51.9% of participating students reported to have had at least one lucid episode in their life (Table 3). As shown in Table 4, the experience with lucid dreaming was significantly related to Age (inline image 44.73, df = 13, P < 0.01) and Secondary School Type (inline image = 16.35, df = 2, P < 0.01). The effect for School Type persisted when we controlled for Age (r = 0.15, df = 544, P < 0.01). Regarding the Age effect, Fig. 3 shows a steady increase of the lucid dreaming experience across years, spanning a range of 52% from age 6 to 19 years. The increase is linear, following the simple equation ‘Frequency of lucid dreaming = p0 + p1 × Age’. Deviations from linearity are non-significant (inline image = 12.16, df = 12), with an ascending slope (p1) of 3.46 (SE = 0.50, P < 0.01) and an ordinate-axis intercept (p0) of 9.34 (SE = 6.36). This linear increase in what we refer to as the lifetime prevalence of lucid dreaming stands in sharp contrast to the distribution of current lucid dreaming (Fig. 4), which demonstrates the need to introduce a temporal reference point when assessing lucid dreaming incidence.

Table 3.   Absolute and relative frequencies of sleep and dream-related variables
Age (n = 215)*Sleep duration mean (SE)Napping % r P
  1. *Sleep duration and napping were recorded in written form only in the second half of the survey (n = 215), in an attempt to better understand the strong age effect we observed in the lucid dreaming data. Two students were not able to specify their usual bedtimes.

610.34 (0.21)14.8Sleep duration and age: −0.73Sleep duration and age: <0.01
710.48 (0.34)27.3Sleep duration and napping: −0.22Sleep duration and napping: <0.01
89.79 (0.25)6.7Age and napping: 0.25Age and napping: <0.01
99.45 (0.36)0.0  
109.61 (0.16)1.5  
119.40 (0.12)2.5  
129.10 (0.14)4.1  
138.99 (0.14)3.2  
148.19 (0.31)6.3  
157.61 (0.19)16.0  
167.61 (0.34)12.1  
177.26 (0.25)13.6  
186.81 (0.20)17.1  
197.66 (0.45)27.8  
Entire sample (n = 622)n %   
Frequency of dream recall
 Never  344.90  
 Almost never  7911.38  
 Sometimes 30043.23  
 Often 18726.95  
 Almost every morning  9413.54  
 Total 694100.00  
Experience with lucid dreaming
 Have had at least one lucid dream in lifetime 360 51.87  
Lucid dreamers only (n = 360)n %   
Frequency of lucid dreams 
 Currently not at all  277.50  
 Very seldom 15242.22  
 Sometimes (min. 1 month−1)  9125.28  
 Often (min. 1 week−1)  7019.44  
 Almost every night  205.55  
 Total 360 100.00  
Lucid last night  5715.83  
Control over dream plot 13336.94  
Table 4.   Inferential statistics on dream-related variables
  1. The influence of school type on dreaming was tested again separately for secondary school types, only (grades 5 and up) because primary school is attended by all children, undifferentiated by intelligence or grade point average. N secondary school type = 310. Univariate analyses for secondary school type controlled for age by defining it as covariate.

  2. *Partial correlation showed this effect to remain significant when controlled for age (r = 0.15, df = 544, P < 0.01).

  3. Partial correlation showed this effect to remain significant when controlled for age (r = 0.16, df = 307, P < 0.01).

Entire sample, n = 622
Frequency of dream recall F df P
  Sex6.081,619<0.05
  Age1.791,619NS
  Secondary school type1.052,540NS
  Sex×school type0.432,540NS
Experience with lucid dreams inline image df P
  Sex0.591NS
  Age24.0111<0.01
  Secondary school type16.352<0.01
Lucid dreamers only, n = 360
 Current frequency of lucid dreams F df P
  Sex0.061,357NS
  Age22.781,357<0.01
  Secondary school type (controlled for age)6.492,303<0.01
  Sex by secondary school type0.122,303NS
 Plot control inline image df P
  Sex1.061NS
  Age28.7913<0.01
  Secondary school type1.272NS
 Recent lucidity (last night) inline image df P
  Sex0.021NS
  Age12.2313NS
  Secondary school type10.822<0.01
Figure 3.

 Percent affirmative responses to the question ‘Have you ever had a lucid dream?’ (ntotal = 622). Error bars correspond to 95% confidence intervals (Wilson). The trend-line represents the significant linear increase of past experience with lucid dreaming across age. Deviations from linearity are non-significant (inline image = 12.16, df = 12), with an ascending slope of 3.46 (SE = 0.50, P < 0.01).

Figure 4.

 Relative prevalence (% of age group) of frequent lucid dreams in students at the time of the interview (n = 360). Please note that this figure relates not to the entire sample but to lucid dreamers only.

As far as the different types of secondary school are concerned, pairwise contrasts between different levels of schooling yielded significant effects for higher level versus both medium and lower level schooling (higher versus lower level: inline image = 14.74, df = 1, P < 0.01; higher versus medium level: inline image = 6.67, df = 1, P < 0.01), but not between lower and medium level secondary school type (lower level versus medium level: inline image = 1.74, df = 1, NS).

Current incidence of lucid dreams (lucid dreamers, only, n = 360)

On a descriptive level, we found that most participating students reported a low incidence rate of lucid dreaming: about 75% of lucid dreamers experience this type of dream once a month or less; only 6% claimed to be lucid almost every night; 8% of lucid dreamers claimed to have been lucid the night prior to the interview. Similar to the results for ‘experience with lucid dreams’, results of a univariate anova using Sex and Secondary School Type as independent variables, and Age as covariate (Table 4) suggest that the current frequency of lucid dreams differs significantly across Age and Secondary School Type (Age: F = 11.37, df = 1, 338, P < 0.01; Secondary School type: F = 6.49, df = 2, 303, P < 0.01).

With respect to Age, we dichotomized responses for better visualization into the categories ‘frequent’ (responses ‘often’ and ‘almost every night’) and ‘infrequent’ (‘sometimes’, ‘very seldom’ and ‘currently not at all’). Frequent lucid dreaming applied to 25% of lucid dreamers (Table 3).

As shown in Fig. 4, frequent lucid dreaming occurs most often before age 17 years. Incidence rates seem to remain at similar levels until age 13 years, after which it steadily declines. Older students (ages 17–19 years) appear to experience lucid dreams only very infrequently. The distribution of current lucidity follows closely that for plot control (Fig. 5). Both distributions are non-linear, plateau-like until age 16 years, and descend sharply from this age on. These distributions are thus more complex than the one for lifetime prevalence of lucid dreaming.

Figure 5.

 Can you usually change or control what happens in a lucid dream? Percent affirmative responses relative to age group (n = 360).

Regarding Secondary School type, multiple post hoc comparisons revealed that lower level school students (mean = 1.28, SE = 0.13) reported significantly less (P < 0.05) frequent lucid dreams than students from either medium (mean = 1.78, SE = 0.12) or higher level (mean = 1.73, SE = 0.07) schools.

Recent lucid dreaming (night prior to interview)

Recent lucidity was related to Type of Secondary School, but not to Age or Sex (Table 4). Not a single student from lower level schools reported to have been lucid very recently, whereas 17% of students from both medium and higher level schools claimed to have been lucid the night prior to the interview.

Control over dream plot (lucid dreamers, only)

Less than half of those who have ever experienced lucid dreams reported to be able to change the dream plot (37%), which suggests that plot control is not as strong a defining characteristic of lucid dreaming as is lucid insight. Control over dream plot differed across age groups (inline image = 28.79, df = 13, P < 0.01), but not between the sexes or school type (Table 4). Not surprisingly, students claiming to have plot control also report a higher lucid dreaming frequency (students with plot control: mean = 1.97, SE = 0.09; without plot control: mean = 1.59, SE = 0.07; t = 3.37, df = 358, P < 0.01).

As shown in Fig. 5, at least partial plot control is most frequent in 7 year olds. It stays at similar levels from ages 9 to 16 years, and drops thereafter. Content-wise, plot control was mostly used to fly or divert violence and aggression directed at the sleeper.

Frequency of dream recall

Most children and young adults remember their dreams at least sometimes (84%), only 5% reported no dream recall at all (Table 3). Univariate anova (Table 4) using Sex and Secondary School Type as fixed factors and Age as covariate showed a significant effect for Sex but not for any other variable. Girls in our sample had slightly but significantly higher recall of their dreams (girls: mean = 2.42, SE = 0.05; boys: mean = 2.24, SE = 0.06). As expected, dream recall was significantly correlated with frequent lucid dreaming (r = 0.14, df = 694, P < 0.01), lifetime prevalence of lucid dreaming (r = 0.10, df = 694, P < 0.01) and recent lucidity (r = 0.13, df = 694, P < 0.01). There was no evidence of a relationship between dream recall and control over dream plot, however (r = 0.06, df = 694, NS).

Sleep-related variables

Univariate anovas were used to test whether dream recall and/or lucid dreaming frequency was influenced by sleep duration or napping. As can be seen from Table 3, we found no evidence of an interrelation between sleep-related and lucid dreaming-related variables. Data indicate a drop in napping at around age 9 years, and a strong increase in students aged 14 years and older (r = 0.25, P < 0.01). This finding is in accordance with the literature (e.g. Thorleifsdottir et al., 2002) and will not be discussed further, as it does not affect results on lucid dreaming. Sleep duration was significantly inversely correlated with age (r = −0.73, P < 0.01) and napping (r = −0.22, P < 0.01); both findings would be expected for this cohort.

To summarize, we found that lucid dreaming is common in young boys and girls, and that it is significantly influenced by age and level of education. At age 6 years, 26% of the students can recall having had at least a single lucid dream in the past. By the age of 19 years, about 78% of students claim to know what it feels like to have a lucid dream. This linear increase in the lifetime prevalence of a lucid dream is not tantamount with the current status of lucid dreaming, however. In fact, our results show that the current incidence of lucid dreams indicates the exact opposite trend. Whereas the lifetime prevalence of lucid dreaming increases, current lucidity decreases with age.

About 15–35% of all 6–16 year olds interviewed (20-60% of 6-16 year old lucid dreamers in the sample, see Figure 4). This incidence rate drops dramatically from that age on. Plot control is more prominent in dreamers who are frequently lucid, and it also strongly decreases with age. Regarding secondary school type, our results show that especially those attending lower level schools experience fewer lucid dreams – both past and current. Lower level school attendance is also associated with reduced plot control in lucid dreams, not, however, with frequency of dream recall. Dream lucidity is moderately related to dream recall, but unrelated to duration of sleep or napping.

Discussion

Results confirm our laboratory observation that lucid dreaming is infrequent and rare in adults. By contrast, lucid dreaming seems to be quite pronounced in young children. Assuming that we have been able to reduce suggestibility and social compliance, it appears as if in published reports of lucid dreaming in adults it is not possible to distinguish between the lifetime prevalence of such dreams and the current status of lucidity. An also surprising result was the relationship of lucid dreaming with cognitive capacity, showing increased lucidity in those attending higher level compared with lower level schools.

Lucid dreaming as exceptional mental state

In the current sample, the very young were the ones with the highest current incidence rate of lucid dreams. Frequency rates remained at similar levels until age 16 years, after which they dropped dramatically. Only one-third of lucid dreamers claimed to be able to change the dream plot, showing that plot control is not automatically activated in lucid dreaming. As in previous reports (e.g. Wolpin et al., 1992), plot control was significantly associated with frequency of lucid dreaming, suggesting that it is susceptible to training. Plot control was also found to vary with age. It remained at relatively high rates (up to 50% of lucid dreams) from 6 to 14 years, and started to decrease from that age on. Lucid dreaming incidence or frequency was not related to sleep duration or napping.

Provided that our results can be trusted, how can these findings be tied into what we already know about sleep and dreaming in children and young adults? Why does lucid dreaming happen and why does it start so early in life?

Based on previous research on lucid dreaming, we are inclined to interpret the current results as evidence that it is an exceptional mental state occurring naturally in the course of brain maturation. The fact that lucid dreaming is more pronounced in students of higher level secondary schools implies that lucid dreaming is linked to the development of cognitive functions. Although we are aware that level of secondary schooling in Germany is not an impeccable indicator of intelligence but also dependent on social and economic factors (e.g. Stubbe et al., 2008), we feel carefully confident in our assumption that lucid dreaming is more readily experienced by those who are more advanced in abstract thinking and reflective insight. Support for this interpretation comes from a study by Lapina et al. (1998). Although details of method and sample characteristics have not been reported, the authors claim a higher level of lucidity in advanced learners. If this is true, however, then why does lucid dreaming decrease in early adulthood, considering that, surely, older students have acquired a higher level of abstraction than younger ones? At this point, we can only speculate about possible and probable causes. One explanation that should be further investigated is that lucid dreaming occurs naturally in the immature but developing brain.

Lucidity would thus be a transient dissociative state during brain maturation that is normally lost in adulthood but still accessible through training.

The gamma band hypothesis

Previous studies (Voss et al., 2009; Hobson and Voss, 2010; Neider et al., 2011) have shown that lucid dreaming is accompanied by increased activity in the gamma frequency band (40 Hz), especially in frontal parts of the brain. Gamma band activity in the 40 Hz range is known to be related to executive ego functions and higher order consciousness (Fries et al., 2007; Paul et al., 2005; Voss et al., 2009). According to the recent literature, people differ in the extent to which they can generate gamma band activity. Furthermore, this capacity varies as a function of developmental brain maturation, which may provide an advantage for such cognitive functions as attention modulation, access to working memory and linguistic development (Giedd et al., 1996, 1999; Gou et al., 2011; Takano and Ogawa, 1998; Uhlhaas et al., 2009, 2010). Only very few studies investigated the link between electroencephalogram gamma band activity and neurophysiological brain maturation processes, such as myelination and pruning. However, analogous to our finding that lucidity rates markedly drop at around age 16 years, Uhlhaas et al. (2009) report that phase synchrony and gamma band power are strongly reduced during late puberty (increasing again in adulthood), marking re-organizational processes of functional networks. Applied to our hypothesis on the relationship between lucid dreaming and gamma band activity, these findings suggest that the two phenomena are linked but that additional (maturational) processes play a role in the ‘natural’ occurrence of lucid dreaming in childhood.

Our survey was a hypothesis-generating pilot study, opening a window for many more important questions that only further research will be able to answer. If, for example, intelligence and maturation processes prepare the brain for dream lucidity, what are the necessary conditions for it to actually occur? The fact that even students of higher level schools are often only infrequently lucid is proof that cognitive ability and brain maturation may be necessary but not sufficient conditions for lucid dreaming in the young.

At present, our working hypothesis is that first episodes of lucid dreaming may be triggered naturally but accidentally. In a still immature system of 40-Hz rhythmicity and frontal lobe integration, gamma band activations must occur not only during waking but sometimes untimely during sleep, as well. If they occurred in non-rapid eye movement (NREM) sleep, these phases would lead to an awakening (Voss et al., in preparation). During REM sleep, they promote lucid dreaming. This means that lucid dreaming may occur during a fluctuation of gamma band activity during the wrong time of the day, which would make it a natural accident of biological rhythmicity.

Methodological critique and outlook

This is admittedly a study of subjectivity using introspection as the main research tool (Meinong, 1923/2010; Voss, 2010), which raises questions on reliability and validity of mnemonic reports, especially of young children and especially on the recollection of dreams. The most important factor limiting interpretation and generalization of results in most surveys on lucid dreaming is social compliance and suggestibility of highly motivated participants (see also Gackenbach, 1991; LaBerge, 1985). Compliant and socially desirable answers are difficult to identify, and we have taken precautions to prevent most of such answers. Aside from excluding all data sets containing affirmative responses to suggestible questions or contradictory statements, interviewers only recorded reports of lucidity if they were corroborated by dream reports and if a proper explanation was provided why the student thought this was a lucid dream. Furthermore, what makes us carefully confident that the remaining students were providing truthful answers was the higher incidence rate of current and past lucidity in the intellectually gifted children. There is no reason to think that intelligence fosters suggestibility, in fact, research on children and young adults suggests the opposite, namely that intelligence is negatively correlated with suggestibility (Chae and Ceci, 2005; Sing and Gudjonsson, 1992).

Comparison with previous findings

Lucid dreaming

The phenomenon of lucid dreaming is often only vaguely understood, even by those who claim to have had lucid dreams in the past. Most studies of lucid dreaming rely on questionable or not properly validated sources (for a review, see Gackenbach, 1991). Our own experience shows that especially online questionnaires, which are currently quite popular, invite unreliable reports that compare meagerly with those collected in the laboratory after REM-awakenings or paper and pencil tests with proper instructions (Voss et al., in preparation). Accordingly, reports of lucid dream experiences in adults (at least once in a lifetime) range from 82% (Schredl and Erlacher, 2004) to 26% (Stepansky et al., 1998). In this survey, we found an overall prevalence rate of 52%, which compares well with a recent representative survey by Schredl and Erlacher (2011), who reported a lifetime prevalence of 51% in adult participants. On the other hand, our group of 18 and 19 year olds had a lifetime prevalence of more than 70%, which shows how important it is to provide age-related incidence rates of lucid dreaming. Another shortcoming of past surveys on lucid dreaming was that they did not distinguish between past and present lucid dreaming incidence. This was one objective of the current study. Our findings strongly suggest that, although most adults can remember having had a lucid dream in the past, they rarely experience them at present.

Miscellaneous findings

Concerning sleep and dreaming, we found no evidence of a confounding effect of sleep duration or napping with regard to either frequency of dream recall or lucid dreaming (Table 5).

Table 5.   Results of univariate anova on the influence of sleep duration and napping on dream-related variables
Frequency of dream recall F df P
  1. Age and sleep duration were defined as covariates, napping and sex as independent variables.

 Sleep duration (h)0.231,209NS
 Napping1.681,209NS
 Sex3.781,209NS
 Age0.351,209NS
 Sex by napping0.391,209NS
Frequency of lucid dreams (lucid dreamers only)
 Sleep duration (h)1.721,209NS
 Napping0.031,209NS
 Sex3.761,209NS
 Age0.431,209NS
 Sex by napping0.911,209NS

We confirmed sex differences in dream recall (Schredl and Reinhard, 2008), but did not observe an age effect. The literature in this regard is controversial, some sources cite a significant decline in older age (Schredl, 2008; Stepansky et al., 1998), while others do not (Schredl and Piel, 2003; Schredl and Reinhard, 2008). Because our age range does not compare with those studies, we refrain from an in-depth discussion at this point. It appears, however, that young children in our current survey were as reliable in their dream reports as their older peers.

Similar to a recent study by Schredl and Erlacher (2011), but to a lesser degree, frequency of dream recall was significantly correlated with frequency of lucid dreams, suggesting that the ability to remember one’s dreams facilitates lucid dreaming or the memory of it.

Clinical implications

Lucid dreaming is of immense value for the study of conscious states. We must ask ourselves, however, what the consequences are for our subjects: will training of lucid dreaming alter cognitive and/or emotional processing capacity in waking? Can lucid dreaming be applied to clinical settings?

What we have observed in the children and young adults of the current survey is what seems like a preparedness for lucid dreaming. Interestingly, plot control was not automatically coupled with lucid dreaming. Does this indicate that they do not recognize the possibility? Is that in turn a function of a Piagetian kind of causality mystery? Do they later learn it? Obviously, it is available to frequent lucid dreamers (Fig. 5), which indicates that plot control must be susceptible to training. Indeed, several of the interviewed young boys and girls stated unasked that lucid dreaming commenced at a time of need when nightmares prevailed.

Narrative 3: (girl, 10): Someone was haunting me. And I was with my girlfriend. The chaser stood before me and wanted to kill me. And then I realized it was only a dream. So I made the person disappear and then suddenly, it wasn’t dark any more.

Students described lucid dreaming in combination with plot control as a sort of self-remedy that helped them not only to sleep through the night but also to achieve a sense of mastery over their emotions. This is especially relevant because plot control was mostly reported in connection with aggression, violence directed at them, or flying. It is especially remarkable because these students had no training and lucid dreaming occurred spontaneously. The present survey did not assess these aspects systematically, but the assertiveness with which students described their experience with lucidity leads us to speculate that lucid dreaming may prove clinically useful in the treatment of children suffering from anxiety disorders or nightmares. Lucid dreaming might even prove useful in abating excessive impulsiveness.

Acknowledgements

The authors thank Katharina Bey, Anna Melnikova, Areta Kempes, Jacqueline Possiel and Jana Speth for conscientiously and enthusiastically conducting the interviews. We thank the following schools for their participation: Gymnasium am Oelberg, Helmholtz-Gymnasium Bonn, Kastanienschule Hennef, Ennertschule Bonn, GHS Oberpleis, Theodor-Litt-Sekundarschule, Korczak-Realschule Troisdorf, Geschwister-Scholl Schule Gemeinschafts Hauptschule Troisdorf, Arnold von Wied Grundschule and Realschule Beuel.

Ancillary