Dreaming in Adolescence

Dreaming in Adolescence by Kelly BulkeleyIn the current issue of the IASD journal Dreaming (Vol. 22, No. 4, pp. 240-252) I have an article with results of a blind word search analysis of a teenage girl’s dream series.  (Many thanks to the anonymous dreamer, “Bea,” and to Bill Domhoff for mediating our interactions.)  The article is my latest effort at developing a method of using statistical patterns in word usage frequency to identify meaningful continuities between dream content and waking life concerns.  I think the results show that we’re making good progress. Here is the abstract of the paper:


“Previous studies of dreaming in adolescence have found that 1) shifts in dream content parallel shifts in cognitive and social development and 2) adolescent girls seem more prone than boys to disturbing dreams and recurrent nightmares.  This paper confirms and extends those findings by using a novel method, blind word searches, to provide results that are more precise, detailed, and objective than those offered by previous studies. The method is used to analyze a series of 223 dreams recorded in a private diary by an American girl, “Bea” (not her real name) from the ages of 14 to 21.  Accurate predictions about continuities between Bea’s dream content and waking life concerns included important aspects of her emotional welfare, daily activities, personal relationships, and cultural life.  The results of this analysis illuminate the multiple ways in which dream content accurately reflects the interests, concerns, and emotional difficulties of an adolescent girl.”

And here are the final two paragraphs:

“These findings underscore an important yet frequently misunderstood point about the continuity hypothesis: The strongest continuities between dreaming and waking relate to emotional concerns rather than external behaviors (Hall and Nordby 1972; Domhoff, Meyer-Gomes, and Schredl, 2005-2006).  Many of Bea’s nightmares do not reflect actual waking experiences, but they do accurately reflect the dire possibilities and worst-case scenarios that trouble her in waking life.  Bea’s nightmares mirror her worries about things that might happen, not necessarily any actual events that have happened.

“For clinicians, therapists, counselors, and teachers who work with adolescents, the Bea series adds new empirical depth to the idea that dreams are meaningful expressions of emotional truth, especially around issues of family history and personal relationships, and perhaps especially for adolescent girls.  It remains to be seen if word search analyses have any further practical value, but the results presented here should certainly encourage anyone who works with teenagers to listen carefully to their dreams for potentially valuable insights into their developmental experiences.”



Dreaming of the Dead: Patterns of Word Usage in Visitation Dreams

Dreaming of the Dead: Patterns of Word Usage in Visitation Dreams by Kelly BulkeleyHere’s a good example of how to use the word search function of the Sleep and Dream Database.   It focuses on “visitation dreams,” i.e., dreams in which people who are dead appear as if alive.  These vivid and highly memorable dreams have been reported in cultures all over the world, in many periods of history.  People today still experience visitation dreams with remarkable frequency (1).   As part of research I’m doing for a new book, I want to learn more about the basic patterns in visitation dreams.  I’m especially interested in their social and emotional aspects.  My hypothesis, based on cross-cultural evidence and the results of a 2007 content analysis study I did of mystical dreams (2), is that visitation dreams tend to be positive experiences, characterized by friendly interactions and low negative emotions.


Can I put that hypothesis to an empirical test?  Can I push the analysis of visitation dreams to a deeper level of detail and identify additional recurrent features?

The SDDb word search function makes this kind of research easier to pursue than ever before.  There’s a new set of dreams in the database, Demographic Survey 2012, which includes a question about visitation dreams.  On the word search page I scrolled down the list of constraint values to harris_2012:Q1030, Visitation Dream, and selected it.  I then selected “Female” from the top line of the constraint value list.  I clicked on “Word Search” again, and entered “25” in the Min Words box under Limit Response Length.  When I clicked “Perform Search” I had a set of 221 reports from women of visitation dreams of 25 or more words in length.  When I repeated this procedure and selected “Male” instead of “Female,” I had a set of 96 reports from men of visitation dreams of 25 or more words in length.

For both the Female and Male sets I searched for all 7 Word Classes and 40 Word Categories, one class or category at a time.  It took about 20 minutes to generate these figures.

The results both confirm and extend my initial hypothesis.  The visitation dreams have many more friendly than physically aggressive social interactions, and generally low proportions of negative emotions (3).  That’s a solid confirmation of previous findings, with some additional details to fill out the picture:

— The overwhelming majority of characters in the visitation dreams are elder family members.  For women, the most frequently used family character words are grandmother, mother, and father.  For men, the most used words are father and dad.

— Other than vision, speech, and some mention of intensity, these dreams have very few other elements of content: low non-visual perception, low colors, low emotions, low cognition, low nature, low non-family characters, low non-friendly social interactions, and virtually no culture references.

This quick exercise in using the SDDb’s word search function has taught me several things.  Visitation dreams do seem to be mostly positive experiences.  They very often include elder family members, i.e. well-known and personally intimate characters with whom the dreamer speaks and has friendly social interactions.  Evidently few other details matter; the dreamer’s focus is squarely on the appearance of the person who is dead but appears as if alive.  There may be some gender differences in which particular family characters show up  most frequently, but the basic patterns of content emerge clearly in both the women’s and men’s reports.

I’m sure I’ll find many more recurrent themes once I read through the dream narratives.  But already, after just a few minutes of statistical analysis, I have a good overview of the dreams that gives me an empirical context for highlighting further subtleties of significance.




(1) In American Dreamers I cite a 2007 survey of 705 American adults that found 38% of the participants had experienced a visitation dream at least once in their lives (p. 32).

(2) “Mystical Dreaming: Patterns in Form, Content, and Meaning” (2009),  Dreaming 19(1): 30-41.

(3) These two sets are not perfect matches for comparison with the SDDb Baselines, since they include all reports of 25+ words, whereas the SDDb Baselines are for reports of 25-49 words and 50-300 words.  But the baselines can still be useful in evaluating the broad patterns of the visitation dreams.

We Have a Winner!

We Have a Winner! by Kelly BulkeleyCongratulations to Michael Paul Coder, the winner of the dream research contest I posed last week.  Michael correctly identified the gender, age, and dream type of six sets of dreams using only word search frequencies and my baseline hypotheses to make his inferences.  He also correctly identified the bonus questions of which sets of dreams came from the same person.  Best of all from my point of view, Michael showed me his work so I could see how he applied the baseline hypotheses to the word search frequencies.  The correct answers are below.


Perhaps it should not be surprising to learn that Michael is a talented and experienced software architect with a website, Lucid-Code, offering a variety of technologies for exploring consciousness in dreaming.


His success is encouraging because it shows that an essentially automatic process of word frequency analysis can reveal meaningful patterns in dreaming. It should give us additional confidence that this kind of approach to dream research can provide quick, accurate, and reliable results.


What other aspects of dream meaning can be illuminated by this method?  That’s an open question….


Set 1: Female, younger than 18, MRDs.

Set 2: Male, older than 18, MRDs.

Set 3: Male, older than 18, MRDs.

Set 4: Male, older than 18, MRDs.

Set 5: Female, younger than 18, MRDs.

Set 6: Male, older than 18, MRDs.

Bonus question: Sets 1 and 5 come from the same person, and Sets 3 and 4 come from the same person.

A Dream Research Contest

A Dream Research Contest by Kelly BulkeleyTo help test the initial hypotheses arising from the SDDb baselines, I am offering a dream research contest, open to anyone willing to give it a try.  The first person to send me (at bulkeleyk@gmail.com) the correct answers will win a $50 gift certificate to Amazon.com.


The contest challenges you to look at the word usage frequencies of six sets of dreams (available here) and make the following three predictions for each set: a) Is the dreamer a male or female? b)  Is the dreamer younger or older than 18? c) Is this a set of most recent dreams (MRDs) or highly memorable dreams (MemDs)?  You will not be able to read any of the dreams–you only get to look at the numerical frequencies of how often they used certain types of words.


I’m going to make it easier by giving three big clues.  These clues are so big, in fact, that they will give you a clear map to reach the correct answers.


Clue #1: The six sets of dreams come from four individual dreamers, let’s call them A, B, C, and D.  Two individuals gave two sets of dreams each.  The other two individuals gave one set of dreams each.  This means the contest has a bonus question: Which sets of dreams come from which individual?


Clue #2: As discussed in previous posts, the SDDb baselines have identified a number of significant differences between male and female dreamers and between MRDs and MemDs.   These differences can serve as analytic tools of comparison for making the kinds of predictions I’m asking you to propose.  Below are several specific tests you can use to identify the dreamer’s gender (male or female) and age (older or younger than 18) and the type of dreams contained in the set (MRDs or MemDs).  (Note: The age tests are not drawn directly from the SDDb but from other sources of research on children’s dreams (e.g., David Foulkes, Calvin Hall).)


Clue #3: There are two cases out of the 18 total predictions (3 questions x 6 sets of dreams) in which the overall conclusion of the tests gives the wrong answer.  In other words, the tests will lead you to the right answer about 89% of the time.  To get 100% of the answers right, you’ll have to add some kind of insight above and beyond the tests.


Good luck!  And please “show your work”–explain to me how you derived your predictions.


Tests for Gender:

If Emotion is higher than the Female MRD baselines –> Female

If Emotion is lower than the Male MRD baselines –> Male

If Fear is higher than the Female MRD baselines –> Female

If Fear is lower than the Male MRD baselines –> Male

If Speech is higher than the Female MRD baselines –> Female

If Speech is Lower than the Male MRD baselines –> Male

If Characters is higher than the Female MRD baselines –> Female

If Characters is lower than the Male MRD baselines –> Male

If Family is higher than the Female MRD baselines –> Female

If Family is lower than the Male MRD baselines –> Male

If Friendliness is higher than the Female MRD baselines –> Female

If Friendliness is lower than the Male MRD baselines –> Male

If Physical Aggression is lower than the Female MRD baselines –> Female

If Physical Aggression is higher than the Male MRD baselines –> Male

If Sexuality is lower than the Female MRD baselines –> Female

If Sexuality is higher than the Male MRD baselines –> Male


Tests for Age:

If Cognition is higher than the MRD baselines –> Older than 18

If Animals is higher than the MRD baselines –> Younger than 18

If Family is higher than the MRD baselines –> Younger than 18


Tests for MRD vs. MemD

If Air is higher than the MemD baselines –> MemD

If Air is lower than the MRD baselines –> MRD

If Flying is higher than the MemD baselines –> MemD

If Flying is lower than the MRD baselines –> MRD

If Family is higher than the MemD baselines –> MemD

If Family is lower than the MRD baselines –> MRD

If Animals is higher than the MemD baselines –> MemD

If Animals is lower than the MRD baselines –> MRD

If Fantastic Beings is higher than the MemD baselines –> MemD

If Fantastic Beings is lower than the MRD baselines –> MRD

If Christianity is higher than the MemD baselines –> MemD

If Christianity is lower than the MRD baselines –> MRD

If Death is higher than the MemD baselines –> MemD

If Death is lower than the MRD baselines –> MRD

The Distinguishing Features of Big Dreams

The Distinguishing Features of Big Dreams by Kelly BulkeleyIf someone presented you with two sets of dreams, one of most recent dreams and one of highly memorable dreams, you could predict with a high degree of confidence which type of dream was in which set, based only on word usage frequencies.

The set with more references to flying, air, family, animals, fantastic beings, Christianity, and death is more likely to consist of highly memorable dreams.

This is a testable hypothesis that emerges out of a comparison of the SDDb baselines for most recent dreams (MRDs) and highly memorable dreams (MemDs), available here.  To be clear, it’s a prediction of probability, not certainty.  Some highly memorable dreams have none of these elements, while some most recent dreams have several of them.  But according to the SDDb baselines, it is the statistical tendency of highly memorable dreams to contain significantly more of these elements than we generally find in most recent dreams.

Whether or not this hypothesis has any practical application, it adds new evidence in support of the theoretical claim that dreams are meaningfully structured not just for the individual dreamer but also in relation to each other.  There really are different types of dreams, and their differences can be expressed in increasingly precise terms.

Other researchers such as Harry Hunt and Don Kuiken have proposed psychological models to account for different types of dreams (what Hunt calls “the multiplicity of dreams”).  I am not yet at the point with the SDDb baselines to feel comfortable engaging directly with their approaches, but that is definitely a long-term goal.

At this stage I want to look more closely at the higher-frequency MemD elements and try to understand what they might contribute to the dreams’ long-term impact on waking awareness.

Flying: Not all dream references to flying involve magical powers–some relate to the flight of birds, or flying on airplanes, or floating in water, or time “flying” by.  But many of the references are indeed about people flying magically, and I think it makes good sense that overall, MemDs have significantly more flying references than MRDs.  I would be surprised if it were otherwise, based on the recurrence of magical flying dreams through cross-cultural history.  Genuine flying dreams tend to be quite vivid and realistic, and it’s reasonable to assume that such unusually stimulating sensations would make a lasting impact on waking awareness.

Air: Some of the air references occur in flying dreams, but in most MemDs the air references appear in different contexts: the dreamer is struggling to breathe, or facing a tornado, or noticing the wind blowing.  I don’t know about dreams of wind, but certainly with dreams of tornados and potential suffocation the memorability of the experience is likely to be very high.  A tornado is the most powerfully destructive form of air in nature, and suffocation is a perennial threat to human life, perhaps especially in sleep for people who snore or have apnea.

Family: References to family members appear often in MRDs; they appear even more often in MemDs.   I think it’s fair to say that most people’s strongest emotional relationships (both positive and negative) are with family members.  Thus it makes sense that their appearance in a dream correlates with high memorability.  Looking in more detail at the word search results, references to parents (e.g., mother, father, mom, dad) tend to be the highest, suggesting that dreams in which the individual is cast as a child or in a child’s role are more likely to be memorable.

Animals: Based on any of several different theories (psychoanalytic, developmental, evolutionary), it could be expected that MemDs would have a higher proportion of animals than MRDs.  Psychoanalytically, animals symbolize powerful instinctual energies. Developmentally, animals appear more often in children’s than in adults’ dreams, and MemDs are often recalled from early in childhood. In evolutionary terms, animals in dreams may reflect ancestral threats that we are innately primed to notice and remember.

Fantastic Beings: This category by definition includes characters who are not “real,” so their appearance in dreams naturally arouses some degree of heightened awareness and emotional impact.  Many of them are perceived as extremely frightening and dangerous to the life of the dreamer.  I was surprised by the SDDb baseline results that the MemDs do not have more fear-related emotions than the MRDs, but perhaps what makes some MemDs different is the supernatural source of the fear. There is a connection to be made here with the notion of “minimally counterintuitive supernatural agents” as used in the cognitive science of religion–dreaming is a rich experiential source of people’s religious and spiritual beliefs about such beings.

Christianity: Many references to Christianity in both MRDs and MemDs are relatively trivial references to Christmas, or mild oaths, or a person’s name.  But more often in the MemDs there are direct references to interactions with Jesus, battles with demons, visiting heaven, and worshipping in church.  In a majority-Christian country like the U.S., where all the SDDb baseline participants reside, this seems like an expectable result.  Insofar as Christianity, like most religions, is concerned with deep questions of morality, suffering, and faith, any dream that refers to religious teachings is likely to register more memorably in the dreamer’s awareness.

Death: Whether considered in religious or secular terms, death surely counts as a major existential concern of human life.  Dreaming itself has long been mythologically associated with death, and cultural traditions all over the world have stories about dreams as a portal to the afterlife.  In MemDs the theme of death takes many forms: other characters dying or being killed right in front of the dreamer, dead relatives appearing as if alive (i.e., visitation dreams), and, more rarely, the dreamer him or herself dying.  When the prospect of mortality arises in a dream, it’s not surprising that the individual takes notice and remembers.

What do these seven higher-frequency MemD elements have in common?

For one thing, several of them involve “counter-factuals,” i.e., phenomena that are literally impossible in ordinary waking life.  Magically flying in the air, encountering fantastic beings, seeing people who are dead appear as if alive–these are strikingly anomalous experiences that stand out from ordinary life and make a big impression on memory.

Secondly, several of the MemD themes involve dire threats to the individual’s life and well-being.  Dreams of death, demons, monsters, wild animals, suffocation, and tornados naturally arouse a host of psychological and physiological responses that can literally seize the dreamer’s attention and hold it long after waking.

Thirdly, a few MemD themes relate closely to the prominent themes of children’s dreams generally, with more animals and higher family references.  As I noted earlier, the SDDb baseline for MemDs includes numerous childhood-era dreams reported by children and adults, so it is definitely skewed toward children’s dream content.  That means the differences between MRDs and MemDs could be explained as artifacts of the differences between adults and children.  I grant there will be a large degree of overlap between highly memorable dreams and children’s dreams–precisely because the most memorable dreams people often recall are dreams from childhood.

Not all children’s dreams are big dreams–but many big dreams are dreams that have been remembered from childhood.






































A Primal Difference (Part 3 of Creating a Baseline for Studying Patterns in Dream Content)

A Primal Difference (Part 3 of Creating a Baseline for Studying Patterns in Dream Content) by Kelly BulkeleyWhat makes unusually memorable dreams different from average, ordinary dreams?  Putting it in Jungian terms, how are big dreams different from little dreams?


The SDDb baselines for most recent dreams (MRDs) and memorable dreams (MemDs) give a very precise and empirically based answer to this question.


MRDs tend to use more words relating to perception, emotion, cognition, social interactions, and culture.


MemD tend to use more words relating to flying, air, family, animals, fantastic beings, Christianity, and death.

Overall, MRDs are more anchored in present waking circumstances, while MemDs seem to have less connection to current social reality.   MRDs reflect more of daily life, while MemDs express deeper existential themes.

These results derive from 828 female MRDs and 691 male MRDs, compared to 801 female MemDs and 504 male MemDs.  You can see the spreadsheet here.

With the help of Dominic Luscinci at Far West Research, I analyzed these four sets of dreams in terms of their similarities and differences, adjusting the levels of statistical significance to account for multiple tests in each word class to protect against type 1 errors.  Fisher’s Exact Test was used in cases where the criteria for chi-square testing were not present.

Before getting into the MRD vs. MemD comparison, I wanted to know what gender differences were most significant. I found that female MRDs and MemDs are more likely than male reports of both types to include references to emotions (especially fear), characters (especially family), speech, and friendly social interactions.  Female MRDs have more perception and cognition, while male MRDs have more physical aggression and sexuality.

There are fewer gender differences in the MemDs than the MRDs.

Then I looked at the MRDs and the MemDs to see what differences show up for both males and females.  I found that MRDs for both genders have more references to emotion, cognition, social interactions, and culture.  MemDs have more references to nature (especially air and flying), characters (family, animals, fantastic beings), Christianity, and death.  The female MemDs have more fire, falling, and physical aggression words.  The male MemDs have more chromatic and achromatic colors.

There are many more differences between MRDs and MemDs than between the males and females.  The comparison with the fewest differences was female MemDs vs. male MemDs; these two sets of dreams were the most like each other.

My first reaction to these findings was surprise that the MRDs had more words relating to perception and emotion, since I expected these indices of intensity and vividness would be more frequent in highly memorable dreams.

But I also felt good because these results basically replicate a 2011 study I did with Ernest Hartmann on big dreams.  In the conclusion of that article we wrote “people’s big dreams are distinguished by a tendency toward ‘primal’ qualities of form and content: more intense imagery, more imagery picturing nightmarish emotions, more nature references, more physical aggression, more family characters, more fantastic/imaginary beings, and more magical happenings, along with less high-order cognition and less connection to ordinary daily surroundings.” (p. 165)

These findings are very similar to the SDDb baseline results.  They give me confidence that these differences between MRDs and MemDs are real and not the result of random variations in the data.

The comparison with the 2011 study is not perfect, since a) that project did not adjust the dream reports for word length, including reports of less than 50 and more than 300 words, unlike the SDDb baselines, b) some aspects of the conclusion (e.g., intense imagery, nightmarish emotions) were derived from Hartmann’s Central Image scoring system and did not emerge from the word search analysis, and c) the participant pool had a big gender imbalance (147 female, 15 male).  However, the mostly female composition of the 2011 study actually points to an even closer alignment with the SDDb female results because the female MemDs (but not the male MemDs) have higher frequencies of fire, falling, and physical aggression, all of which seem consistent with the 2011 study’s conclusion.

In a future post I will look at the SDDb’s high-frequency MemD elements–flying, air, family, animals, fantastic beings, Christianity, death–to try and discern what each of them adds to the dream’s memorability and impact on the dreamer.