The Dreams of a Sight-Impaired Person

The Dreams of a Sight-Impaired Person by Kelly BulkeleyA study from 2015 about the distinctive patterns in the dreams of “Jasmine,” a young sight-impaired musician.  First presented at the annual conference of the International Association for the Study of Dreams, in Virginia Beach, Virginia.


This presentation demonstrates the use of a digital word search method to study 800 dreams recorded in a personal journal over a period of several years by a female participant.  A “blind” analysis, focusing exclusively on the frequencies of word usage in her dreams, enabled several accurate predictions about her waking life, including the fact that she has been sight-impaired since early childhood.


The word search method I will describe has relevance for researchers from any background who are interested in the systematic study of dream content.  The great advantage of quantitative approaches to the study of dreams, like the Hall and Van de Castle system, is that they provide objective statistical results that other researchers can compare, replicate, and extend.  However, there are serious disadvantages to traditional hand-coding methods—they are time-consuming, labor-intensive, hard to learn, and vulnerable to problems with inter-coder reliability.  Particularly when the coding systems being employed are untested or idiosyncratic, the results can be disappointing.  Digital word search methods provide a viable alternative: they are fast, easy to use and share, and reliably consistent in their results.  Of course, such methods do not supplant other approaches to dream research.  Digital tools are best used as complementary resources in the study of dreams, providing an empirically grounded assessment of basic patterns of dream content.

Over the past several years, and usually in collaboration with G. William Domhoff and Adam Schneider, I have pursued several studies using word search methods to analyze particular sets or series of dreams:

The Religious Content of Dreams: A New Scientific Foundation. Pastoral Psychology 58(2): 93-101.

[Merri, Barb Sanders]

Seeking Patterns in Dream Content: A Systematic Approach to Word Searches. Consciousness and Cognition 18: 905-916.

[Paul, HVDC Male and Female Norms]

Detecting Meaning in Dream Reports: An Extension of a Word Search Approach. (co-authored with G. William Domhoff).  Dreaming 20(2): 77-95.


Dreaming in Adolescence: A “Blind” Word Search of a Teenage Girl’s Dream Series. Dreaming 22(4): 240-252.


Digital Dream Analysis: A Revised Method. Consciousness and Cognition 29: 159-170.

[The Engine Man, Barb Sanders, HVDC Male and Female Norms]

All of these studies, and several others not yet published, have led to new insights about using word search methods to identify meaningful patterns in dream content.  I have focused special attention on projects using a “blind analysis” framework in which I start with as little information as possible about the dreamers, focusing exclusively on the statistical frequencies of word usage in their dreams.  I make a point of knowing nothing about the dreamers’ personal lives, and I do not read the narratives of their dreams.  This approach imposes an artificial set of constraints that concentrates all my analytic attention on the word usage frequencies.  By bracketing out every other source of information, I try to glean as many insights as possible from the word usage patterns alone.  The present study is the latest in this series of word search investigations.


“Jasmine” (not her real name) contacted Domhoff in 2014 and offered to share her lengthy dream journal with him for research purposes.  He then contacted me, and with her permission we arranged a blind analysis process whereby I uploaded four sets of Jasmine’s dream reports into the Sleep and Dream Database (SDDb) in order to perform the subsequent word searches.  I knew nothing about Jasmine besides her being female, and my only other bit of knowledge was that the four different sets of her dreams followed a chronological order.


Once the dream reports were uploaded, I used the SDDb’s 2.0 word search template to calculate the word usage frequencies for each of 40 categories (many of which overlap with Hall and Van de Castle categories), for each of the four sets of dreams.  I then compared the results with the SDDb Baselines, a large collection of “most recent” dreams from various sources that provides a measuring stick for looking at how a given set of dreams relates to what current evidence suggests are the normal patterns of ordinary dream content.  I also compared the four sets of dreams with each other, looking for patterns of consistency and variation from one set to the next.

Based on these word search findings, I formulated 21 inferences about Jasmine’s waking life.  These inferences derived from the continuity hypothesis that patterns of dream content accurately reflect a person’s concerns, relationships, and activities in the waking world.  I used as a working assumption the idea, supported by previous studies, that frequencies of word usage in dreams are continuous to some degree with the individual’s waking life concerns.  My inferences about Jasmine’s waking life were predicated on that assumption.

I sent the inferences to Domhoff, who forwarded them to Jasmine.  Once I received her responses, I was no longer “blind,” in the sense that now I knew a great deal about her personal life and could see where her waking life concerns had crossed over into her dreaming.  We engaged in several more email exchanges to answer her questions and learn more about various details of her personal upbringing and family background.


Appendix 1 presents the results of the word search analysis using the SDDb 2.0 template.  Below are the 21 inferences I formulated based on the word search results, and Jasmine’s initial responses to them (in bold italics):

1. Jasmine is blind or sight-impaired. yes

2. Hearing is an especially important sense for her. yes

3. Jasmine has very close relations with her family.

Well, it’s complicated.  I would say that I’m close to some of them, particularly on my mom’s side.  My siblings live far away, and so sometimes I miss them and remember things we used to do together.  I have complicated relationships with those on my dad’s side in general. 

4. Her most important relationship is with her mother. Yes!

5. She is very involved with birds, especially chickens.

Yes, I had pet chickens almost my whole life growing up, and other feathered pets as well. J 

6. She enjoys books, reading, and writing. Yes

7. She is very active musically, especially singing and playing

piano.    Yes!

8. She is, or recently has been, a student in college. yes

9. She was raised as a Christian and she now regularly attends church.

Well, this is complicated too.  My family wasn’t really that religious, but sometimes we would go to church when I was little. 

Domhoff also asked that, in addition to making general inferences, I make some specific inferences based on comparing the first two sets to the third set, and the third set to the fourth set.

Looking at the time period covered by the third set compared to the earlier two sets of dreams, Jasmine was:

10. More fearful and anxious. Not sure, I can’t really remember.

11. But also happier than before.

           This is possible.  I started to gain more independence in this period, so I think 10 and 11 might both be true, to a degree. 

12. Less involved with her family. Yes, this jives with the above. J

13. Drinking more coffee. I don’t remember starting then, but I like coffee now.

14. More involved with piano music. yes

15. More involved with religion and church. Yes, because I started playing the organ. 

Looking at the time period covered by the fourth set of dreams compared to the time period covered by the third set, Jasmine was:

16. Less fearful and anxious. Not sure

17. More involved with her family. No, not really.

18. More involved with birds.

No, I didn’t have any pets, and my last chicken and my cockatiel died before this time.  This didn’t really happen, but I can remember dreaming a lot of dreams centering around forgetting to feed, or otherwise take care of them. 

19. More actively engaged with the public world (indicated by more references to movement, travel, work, clothing).   Yes, we could say that I gained even more independence at this time. This is when I got my first real job!

20. Still very musically active with piano and singing. yes

21. Even more involved with a Christian church. Yes, playing the organ and attending regularly, but not always seeing things like they do. 



Of the 21 dream-based inferences about her waking life, Jasmine confirmed 15 of them as accurate (and 9 of 9 for the general inferences at the beginning).  She said she was not sure about four of the inferences, although she allowed that two of those “might both be true, to a degree.”  That would raise the total number of correct inferences to 17 out of 21. Either way, this is an accuracy ratio comparable to previous studies of the Van series (12 of 14 inferences correct) and the Bea series (11 of 15 inferences correct).

On the issue of her blindness, I based my inference on the fact that her frequency of vision-related words was lower than her frequency of hearing-related words.  That is a very rare pattern that I have only found in one other group—the blind people whose dreams Hurovitz and his colleagues wrote about in their 1999 study.  I’m not sure this will hold up as a universally valid principle, but in Jasmine’s case the relatively low frequency of vision-related words combined with the high frequencies of other sensory terms made it led me to infer that she was blind or sight-impaired in waking life.

In their 1999 study Hurovitz et al. found some evidence of continuity between blind people’s difficulties moving around in waking life and their dreams of movement and transportation.  I did not think of that when formulating my inferences, but now that I look at the findings with Jasmine I do see elevated levels of references to walking & running, falling, and transportation, compared to the frequencies in the SDDb baselines.

The accurate inferences about her family and her mother underscore the strong continuities between waking and dreaming in terms of social relationships.  This is one of the areas where dream content most directly mirrors the feelings and concerns of a person’s waking life.  Further evidence of the same point comes with the correct inference about chickens.  Such a high frequency of words relating to chickens and other birds is quite rare, and in this case it accurately reflects the emotional importance of her relationships with these creatures over the course of her life.

Jasmine’s extremely high frequency of references to art-related words involving music, combined with her high frequency of hearing words, suggested that music plays a big role in her waking life, as indeed it does.  It turns out that Jasmine is quite gifted musically; she sings, plays piano, and in recent years has been playing the organ at churches.  Her religious activities have in fact increased in recent years, as I inferred, but she expressed some discomfort with the doctrinal teachings at these churches.

The inferences about changes in her life from one period of time to another were less successful than the inferences about her waking life in general.  Jasmine disagreed with two of my inferences predicting that between the third and fourth sets she became “more involved with” her family and with birds.  One possible explanation for the mistaken inference is that I used the word “involved” ambiguously in the question.  Jasmine seemed to take it as asking whether or not she had more direct physical interactions with her family and birds.  To that question she answered no, she was not more involved with them in that literal sense, in large part because she had just started a new job and was away from her usual home setting more than ever before.  However, if I had phrased the inference to include not only “physical interactions with” but also “thoughts or emotional concerns about,” she may have responded differently.  At this key transitional time in her personal development, the rise in dream references may have reflected a greater emotional concern about her relations with family and birds, even though she was not actually interacting with them as much as in the past.  As Jasmine herself said in her response, her dreams at this time had many instances of worrying about her birds, despite no longer having any actual birds in her life.

A similar phenomenon arose in the study of Bea’s dreams, when she dreamed more often of her family and her pet dogs when she left home to go away to school.  Here’s a quote from the Bea paper: “In college Bea was physically present with her parents less than before, but she was thinking and worrying about them more than ever. Her anxious feelings, not her physical interactions, carried over into her horrible dreaming.” (248)  The two mistaken inferences I made in analyzing Jasmine’s dreams may, in retrospect, be consistent with what I found in the Bea series, namely that “dreams accurately reflect emotional concerns but not necessarily actual events.” (248)  They are more like a poetry journal than a newspaper article.


There are several limitations to this study.  Jasmine’s dream reports were unusually long, so it’s unclear to what extent we can compare her word usage frequencies with those gathered from shorter dreams.  There may have been demand effects in terms of influencing Jasmine to give us socially pleasing answers (although she did disagree with two of my inferences and expressed uncertainty about four others).  The word search method does not distinguish between literal and metaphorical uses of words, and the analysis is vulnerable to distortions from dream reports that include spelling mistakes, non-dream associations, and extraneous comments.  These limits add a note of caution to any conclusions drawn from this approach.


This study of Jasmine’s dreams has shown that a blind word search method can accurately identify many important concerns in a person’s waking life, including her relationships, interests, activities, and personal attributes.   The results add evidence to support of the continuity hypothesis, and they shed new light on patterns in dream content among sight-impaired people.

Looking at the findings of this study in the context of the other word search projects mentioned earlier, I would like to close with the two observations.

First, these methods are helping to lower the cost of entry for innovative dream research.  Databases like the SDDb and Domhoff and Schneider’s DreamBank offer free and open access to large collections of dreams from various sources, along with powerful tools of word searching and statistical analysis.  In coming years there will hopefully be more online resources like these, with even better analytic tools.  This bodes well for our field in terms of enhancing accessibility and stimulating creativity among scholars from various backgrounds.

Second, blind analysis studies are the incubators of proto-algorithms for future technologies of digitized dream interpretation.  What I am doing when I perform a blind analysis is essentially the same as what a well-designed computer program could potentially do when presented with the same information.  My inferences are attempts, using high-quality data and systematic methods of analysis, to identify the highest-probability correlations between dream content frequencies and waking life concerns.  The more of these correlations we identify using blind analysis and other methods, the better those interpretive algorithms of the future will be.


Note: this paper has been posted on as SDDb Research Papers #4.

The accompanying appendix has been posted here.



The Science of Dreaming: 9 Key Points

The Science of Dreaming: 9 Key Points by Kelly BulkeleyThe most important findings of scientific dream research can be summarized in nine key points.  Many important questions about dreaming remain unanswered, but these nine findings have solid empirical evidence to support them. 

  1. Rapid eye movement (REM) sleep is a trigger for dreaming, but is not identical with dreaming. All mammals have sleep cycles in which their brains pass through various stages of REM and non-REM sleep.  Dreaming seems to occur most often, and most intensely, in REM sleep, a time when many of the brain’s neuro-electrical systems have risen to peak levels of activation, as high as levels found in waking consciousness.  However, dreaming occurs outside of REM sleep, too, so the two are not identical; REM sleep is neither necessary nor sufficient for dreaming.
  2. REM helps the brain grow. The fact that REM sleep ratios are at their highest early in childhood (newborns spend up to 80% of their sleep in REM, whereas adults usually have 20-25% of their sleep in REM) suggests that REM, and perhaps dreaming, have a role in neural maturation and psychological development.
  3. Dreaming also occurs during hypnogogic, hypnopompic, and non-REM stage 2 phases of sleep. In the transitional times when a person is falling asleep (hypnogogic) or waking up (hypnopompic), various kinds of dream experiences can occur.  The same is true during the end of a normal night’s sleep cycle, when a person’s brain is alternating exclusively between REM and non-REM stage 2 phases of sleep, with a relatively high degree of brain activation throughout.  Dreams from REM and non-REM stage 2 are difficult to distinguish at these times.
  4. The neuro-anatomical profile of REM sleep supports the experience of intense visionary imagery in dreaming. During REM sleep, when most but not all dreaming occurs, the human brain shifts into a different mode of regional activation.  Areas of the prefrontal cortex involved in focused attention and rational thought become less active, while areas in the limbic system (involved in emotional processing, memory, and instinctive responses) and the occipital lobe (involved in visual imagination) become much more active.  This suggests that the human brain is not only capable of generating intense visionary experiences in dreaming, it has been primed to do so on a regular basis.
  5. The recurrent patterns of dream content are often continuous with people’s concerns, activities, and beliefs in waking life. This is known as the “continuity” hypothesis, and it highlights the deep consistency of waking and dreaming modes of thought.  People’s dreams tend to reflect the people and things they most care about in the waking world.  A great deal of dream content involves familiar people, places, and activities in the individual’s waking life.  The dreaming imagination is fully capable of portraying normal, realistic scenarios. This means dreaming is clearly not a process characterized by total incoherence, irrationality, or bizarreness.
  6. The discontinuities of dreaming, when things happen that do not correspond to a normal waking life concern, can signal the emergence of metaphorical insights. Research on the improbable, unreal, and extraordinary elements of dream content has shown that, on closer analysis, this material often has a figurative or metaphorical relationship to the dreamer’s waking life.  Metaphorical themes and images in dreams have a long history in the realm of art and creativity, and current scientific research highlights the dynamic, unpredictable nature of dreaming as an endless generator of conceptual novelty and innovation.
  7. Dream recall is variable. Most people remember one to two dreams per week, although the memories often fade quickly if the dreams are not recorded in a journal.  On average, younger people tend to remember more dreams than older people, and women more than men.  Even people who rarely remember their dreams can often recall one or two unusual dreams from their lives, dreams with so much intensity and vividness they cannot be forgotten.  Dream recall tends to respond to waking interest.  The more people pay attention to their dreams, the more dreams they are likely to remember.
  8. Dreaming helps the mind to process information from waking life, especially experiences with a strong emotional charge. From a cognitive psychological perspective, dreaming functions to help the mind adapt to the external environment by evaluating perceptions, regulating emotional arousal, and rehearsing behavioral responses.  Dreaming is like a psychological thermostat, pre-set to keep us healthy, balanced, and ready to react to both threats and opportunities in the waking world. Post-traumatic nightmares show what happens when an experience is too intense and painful to process in a normal way, knocking the whole system out of balance.
  9. The mind is capable of metacognition in dreaming, including lucid self-awareness. During sleep and dreaming the mind engages in many of the activities most associated with waking consciousness: reasoning, comparing, remembering, deciding, and monitoring one’s own thoughts and feelings. Lucid dreaming is one clear example of this, and so are dreams of watching oneself from an outside perspective.  These kinds of metacognitive (thinking about thinking) functions were once thought to be impossible in dreaming, but current research has proven otherwise.  Dreaming has available the full range of the mind’s metacognitive powers, although in different combinations from those typically active in ordinary waking consciousness.

The Science of Dreaming: 9 Key Points by Kelly Bulkeley

For further reading:

Barrett, Deirdre and Patrick McNamara, ed.s.  The New Science of Dreaming.  Westport: ABC-Clio, 2007.

Bulkeley, Kelly.  Big Dreams: The Science of Dreaming and the Origins of Religion.  New York: Oxford University Press, 2016.

Domhoff, G. William.  Finding Meaning in Dreams: A Quantitative Approach.  New York: Plenum, 1996.

Hurd, Ryan and Kelly Bulkeley, ed.s.  Lucid Dreaming: New Perspectives on Consciousness in Sleep.  Westport: ABC-Clio, 2014.

Kryger, Meir H., Thomas Roth, and William C. Dement, ed.s. Principles and Practice of Sleep Medicine.  Fourth Edition.  Philadelphia: Elsevier Saunders, 2005.

Maquet, Pierre, Carlyle Smith, and Robert Stickgold, ed.s.  Sleep and Brain Plasticity. New York: Oxford University Press, 2003.

Pace-Schott, Edward, Mark Solms, Mark Blagrove, and Stevan Harnad, ed.s.  Sleep and Dreaming: Scientific Advances and Reconsiderations. Cambridge University Press, 2003.

Pagel, James.  The Limits of Dream: A Scientific Exploration of the Mind/Brain Interface. New York: Academic Press, 2010.

Solms, Mark.  The Neuropsychology of Dreams: A Clinico-Anatomical Study.  Mahway: Lawrence Erlbaum, 1997.



1,001 Nights of Dream Recall

1,001 Nights of Dream Recall by Kelly BulkeleyBetween January 8, 2015 and October 4, 2017, I remembered and recorded a dream every night for 1,001 consecutive nights.  Now I’m studying the dreams and trying to find insights that can help in exploring the dream series of other people.  I don’t expect anyone to accept my personal dreams as conclusive evidence for any general theory of human dreaming.  Instead, I offer them as way of being transparent about the experiential grounding of my research pursuits.  This is one of the ways I get ideas for new projects.

All of the dreams are available online for further study in the SDDb, in the “Sample Data” section.

From a scientific perspective, the value of an introspective project such as this is to generate working hypotheses for future studies.  Trying to study another person’s long diary of dreams can be very challenging, especially at the outset when the researcher is facing a huge mass of texts with multiple dimensions of meaning.  I appreciate anything that can provide some initial orientation and help to steer the direction of the analysis.  By studying my own dreams, which I know from both a first- and third-person perspective, I can quickly and easily identify some patterns of meaning that seem worth further exploration.   Maybe they will apply to someone else’s dream series, maybe they won’t; either way, it helps the analytic process get going.

Remembering and Recording the Dreams

The method I use for keeping my dream journal is fairly typical.  I keep a pad of paper and a pen by my bedside, and when I wake up in the morning I immediately write down whatever dreams I can remember, before getting out of bed or turning on the light.  Then later in the morning I type the dream into a digital file, along with associations, memories, and thoughts about what the dream might mean.

During the three years of recording this series of 1,001 dreams, I did my best to wake up slowly each morning, so the images and feelings from the preceding dreams could coalesce in my memory.  I don’t believe I dreamed more during this time than I did in previous periods of my life; rather, I devoted more energy to remembering the wispier, more evanescent kinds of dreams that, in previous years, had not crossed the threshold into waking memory.  I made a more determined effort to protect the space around the transition from sleeping to waking, even during circumstances when that was difficult to do (e.g., on international plane flights, during family holidays).  Often it took a few moments of quietly lying in bed with my eyes closed before the vague feeling “I know I was just dreaming,” could eventually take form into a specific memory of what I was just dreaming about.  Often there were “aha!” moments when suddenly a whole long dream came back to me, which I surely would have forgotten if I had immediately leapt out of bed upon awakening.

I also put more conscious intention into the other end of the transition, from waking into sleeping, as I carefully set up my journal and pen each night before turning out the light.  I did not set specific dream incubation questions during this time, but simply tried to signal to myself that I was ready and willing to record whatever dreams might come during that night’s sleep.

These were not extreme or burdensome behaviors; they required consistency of purpose, but no heroic feats of will.  I never set any long-term goals or numerical targets.  Instead, I just focused on each new night and each new dream, figuring the time would come when I could survey the series from a broader perspective.

About a month ago I finally did the math, and realized that October 4th would mark 1,001 nights of dream recall in a row, an enchanting milestone.  This seemed like a large enough collection of dreams to pause, look back, and see what I could learn.

Patterns of Word Usage

The reports comprise a total of 93,050 words, with an average length per dream report of 93 words, and a median length of 73 words.  The shortest dream in the series has 9 words, and the longest has 728 words.  The average length of these dreams is not unusual, compared to other people whose dreams have been analyzed in this way.  Some people have much longer dreams than I do, and some people have much shorter dreams.  This series of 1,001 dreams, then, includes mostly dreams of middling length.

To highlight the patterns and themes in a series like this, I start by comparing it to what I call the “SDDb baselines,” two large collections of male (N=2,135) and female (N=3,110) dreams that have been systematically gathered and analyzed using a template of 8 classes and 40 categories of word usage.   I use the baselines a measuring stick for identifying possible continuities and discontinuities between the dreams and the individual’s waking life.

The results of this comparative analysis are presented in an accompanying spreadsheet, “1001 Nights Data.”  Some of the discussion below draws on an earlier analysis I wrote about an overlapping set of my dreams.

In relation to the SDDb baselines (an average of 100 words per report for the females and 105 for the males), my dreams are a little shorter than average (93 words per report).  The results for each of the 8 classes of word usage are summarized below.  Compared to the male and female baselines, my 1,001 dreams have:

  1. Perception: More references to vision and colors.
  2. Emotion: Many more references to wonder/confusion, and more to happiness.
  3. Characters: Fewer references to family characters (although the word “wife” is mentioned very frequently), more references to animals (especially cats), and slightly more references to females than males.
  4. Social Interactions: Slightly more references to sexuality.
  5. Movement: Fewer references to death.
  6. Cognition: More references to thought, fewer to speech.
  7. Culture: Fewer references to school, food/drink, religion, somewhat more to sports (especially baseball and basketball).
  8. Elements: More references to water, somewhat more to earth.

These findings provide the basis for a “blind analysis,” which means making predictions about continuities between these patterns of word usage in dreaming and the individual’s waking life activities, beliefs, and concerns.  If I pretend I knew nothing about the dreamer of these 1,001 dreams, and I only had these word usage frequencies to consider, I would infer this individual:

  1. Is visually oriented
  2. Often experiences wonder/confusion
  3. Is relatively happy
  4. Is married
  5. Cares about cats
  6. Has fairly equal relations with men and women
  7. Is sexually active
  8. Is not concerned about death
  9. Is not highly verbal
  10. Is not highly involved with schools
  11. Is not highly concerned about food/drink
  12. Is not highly concerned about religion
  13. Has lots of interactions with water and earth

1,001 Nights of Dream Recall by Kelly BulkeleyMost of these inferences—I’d say 11 of 13—are unmistakably accurate in identifying a continuity between a pattern of dream content and an aspect of my waking life concerns.  The two I would question are numbers 10 and 12.  Regarding the low frequency of dream references to school, I do in fact engage in a great deal of teaching and educational work, but it’s almost entirely online, and I rarely set foot inside a traditional school any more.  Also, I no longer have school-age children living at home.  So it seems my dreams are continuous with my physical behaviors relating to schools, but not with my computer-mediated educational activities.

Regarding the low frequency of religion references, I most certainly do have great interest in religion, going back to my masters and doctoral studies at the Divinity Schools of Harvard and University of Chicago.  So the inference seems very wrong at this level.  And yet, at another level it seems more accurate.  I was not raised in a religious household, I do not personally identify with any official religious tradition, and I rarely attend religious worship services.  Compared to other people I’ve studied with very high frequencies of references to religion in their dreams, I am a much less personally pious person.  Perhaps what this suggests is that the dreams are accurately reflecting the fact that religion may be an important intellectual category for me, but it is not a personal concern.  My spiritual pursuits are more likely to be expressed in dreams with references to other word categories like water, art, sexuality, animals, and flying.

Shorter versus Longer Dreams

Earlier this year I looked at different set of my dreams to get some idea about possible differences between shorter and longer dreams.  This question rose in relevance when I realized, as noted earlier, that my increased recall seemed to depend in part on the recollection of relatively shorter dreams that in the past I did not fully remember or write down.

There were two main findings of that earlier study.  First, most of the patterns in content appeared in dreams of all lengths, from the shortest (less than 50 words per report) to the longest (more than 150 words per report).  Here’s a summary of what I found:

“The results of this analysis suggest that shorter dreams are not dramatically different from longer dreams in terms of the relative proportions of their word usage.  The raw percentages of word usage do rise from shorter to longer dreams, of course, but the relative proportions generally do not.”

Second, the longer dreams did have proportionally more references to a few word categories, chiefly Fear, Speech, Walking/Running, and Transportation. Another quote:

“These are the word categories that seem to be over-represented in longer dreams.  They are significant contributors to what makes long dreams so long.”

Returning to the present collection of 1,001 dreams, I divided the series at the median point into two groups: the shorter dreams (72 words or less, 500 reports total) and the longer dreams (73 words or more, 501 reports total).  I used the same SDDb word searching template with each of the two groups as I used with the full series, and then I compared their frequencies of word usage.  The biggest variations between the shorter and longer dreams appeared in the following categories:

  • Touch
  • Fear
  • Anger
  • Physical aggression
  • Walking/Running
  • Speech
  • Transportation
  • Water

This list adds a few other categories that may be characteristic of longer dreams.  Each of these categories has a dynamic quality.  Touch is a physical interaction.  Fear and anger are strong and unpredictable emotions, usually prompted by something in the external environment.  Physical aggression combines the previous categories (touch, fear, anger) and possibly intensifies them.  Walking/Running and Transportation both involve physical movement from one place to another.  Speech implies a context of interpersonal communication, people talking with each other.  Water, the “universal solvent,” is ever-shifting in its states (gas, liquid, solid) and its movement through human life.

When these elements appear in my dreams they seem to have the effect of expanding the range of experience, stimulating more interactions, and lengthening the narrative.


I don’t know if any of this applies to anyone else’s dreams.  I do, though, believe that several of the insights gained here can provide working hypotheses for studying other series of dreams.  I will be keeping these ideas in mind as I explore new dream series:

  • The recall and recording methods of the dreamer influence the types of dreams included in the series.
  • Personal relationships are an area of especially strong continuity between waking and dreaming life.
  • The use of religion-related words in dreams may be discontinuous with spiritual interests in waking life.
  • Shorter dreams have mostly the same general proportions and patterns of content as found in longer dreams.
  • Longer dreams tend to include more dynamic elements.





The accompanying spreadsheet can be found here:

More description of the SDDb baselines can be found in Big Dreams: The Science of Dreaming and the Origins of Religion (New York: Oxford University Press, 2016).

The earlier study of short versus long dreams can be found here:

Short vs. Long Dreams: Are There Any Differences in Content?


A Guide to the Sleep and Dream Database

A Guide to the Sleep and Dream Database by Kelly BulkeleyJust added to the SDDb library is A Guide to the Sleep and Dream Database, a text I hope will help to make the resources of the site more accessible to teachers, students, researchers, and anyone interested in dreams.

The guide is the first in a series of what I’m titling SDDb Research Papers.  This series will present new findings from projects that use the dream materials and analytic tools of the SDDb.  The papers will share works-in-progress in the empirical study of dreams.

Here is the abstract for the first paper:

Intended for newcomers to dream research, this guide offers an introduction to the functions of the Sleep and Dream Database (SDDb), an open-access digital archive that includes tens of thousands of dream reports, along with survey data about sleep, dreaming, and demographic variables.  Readers are shown how to use the Survey Analysis and Word Searching functions of the SDDb to study a variety of questions about dreaming.  The topics discussed here as illustrations include gender and age variables in dream recall; differences between men and women in the frequency of fear in their dreams; and the meaningful patterns of content in a woman’s long-term dream journal.

A Guide to the Sleep and Dream Database by Kelly Bulkeley

Dreams Conference 2017


Dreams Conference 2017 by Kelly BulkeleyThe 34th Annual Conference of the International Association for the Study of Dreams begins on Friday, June 16, in Anaheim, California.  I’ll be making several presentations, all of which draw on findings from the Sleep and Dream Database (SDDb).

On Saturday, a panel on “Dreams and Current Politics” will include Jody Grundy, Jeremy Taylor, and myself, talking from various perspectives about dreams as reflections of people’s political beliefs.  My presentation will describe a demographic survey of 2,285 American adults.  The initial findings correspond to previous studies I’ve done on this topic: political conservatives tend to sleep better (less insomnia) and remember fewer dreams than liberals, who tend to sleep worse and remember more of their dreams.

The survey also revealed a troubling divide on the demographic questions of personal income and education.  Wealthier people tend to sleep better and remember more dreams than poor people.  Likewise, more educated people tend to sleep better and remember more dreams than less educated people.

I’m looking forward to a lively discussion with the other panelists and the audience about the significance and implications of these results.

Dreams Conference 2017 by Kelly BulkeleyOn Sunday, a session on “Dreaming and Waking Continuities” will include Nori Muster, Jayne Gackenbach, and myself, all focusing in various ways on empirical methods of identifying clear and meaningful continuities between dream content and waking life concerns.  My presentation will describe the results of analyzing the long-term dream journals of four women, using the word search tools in the SDDb.   I’ll show the audience how to practice the method of “blind analysis,” an interpretive approach in which the word usage frequencies of a person’s dreams are used to make inferences about the person’s most important concerns, activities, and relationships in waking life.  Hopefully the audience will gain a practical, hands-on appreciation for what this method can and cannot achieve.

On Monday I’m part of a panel titled “Research 101” with Justina Lasley, Tracey Kahan, Jayne Gackenbach, Bob Hoss, and Michael Schredl.  The plan is for each of us to take 5-10 minutes to describe our research and the main findings that we think anyone interested in dreams should know about.  It’s an ambitious program, and likely to put on display both the agreements and disagreements that define the field today.

Dreams Conference 2017 by Kelly BulkeleyFor my portion of the panel, I will draw upon what I did at the “OMSI After Dark” event earlier this year at the Oregon Museum of Science and Industry in Portland.  It was the first time I gave a public show-and-tell presentation about the SDDb, and I learned a lot about how to engage ordinary people in the most intriguing aspects of dream research.

Last but not least, on Monday evening I have the honor of introducing one of the conference’s keynote speakers, Professor G. William Domhoff of the University of California, Santa Cruz.  Prof. Domhoff (Bill) has been a tremendous supporter and guide in my efforts to develop and improve the SDDb.  He and his colleague Adam Schneider have done revolutionary work in bringing the Hall and Van de Castle coding system into the digital era, and I am very excited to hear Bill’s talk, which is titled “Seven Surprising Discoveries That Changed My Thinking About Dreams.”  I’ve got some guesses about what they are…

Short vs. Long Dreams: Are There Any Differences in Content?

A word search analysis of a five-year selection from my own dream journal reveals the same consistent patterns of content in both shorter and longer reports.

I’ve been wondering about this question for a long time now.  Are shorter dreams different in any fundamental way from longer dreams?  Some people naturally remember only brief dream fragments and images, while other people can remember extremely elaborate and detailed dream scenarios.  Most researchers prefer to analyze reports in the “Goldilocks zone,” not too short or too long, just right in the middle.  That is a reasonable methodological choice, but it still leaves unanswered the question I’ve been pondering.

To continue developing the word search tools of the Sleep and Dream Database (SDDb), I really need to get some clarity on this point.  The frequencies of word usage identified by the SDDb tools vary a great deal depending on whether the dreams have a smaller or larger number of total words.  Is this a problem, or not?

I also have a personal reason for wanting to explore the question.  In early 2015 I began a new approach to my own dream journaling practice, which has led to at least one remembered dream every night for more than two years.  This is approximately double my recall rate for the previous several years.  The 2015 dreams were also shorter on average (74 words) than the dreams from previous years (all averaging 100+ words).  This made me wonder about possible changes in the content patterns of my dreams before and after 2015.

With all of this in mind, I started by tabulating the distribution of my dreams over a five-year period of time (2012-2016), separating them into four categories of word length (less than 50 words, 50-99 words, 100-149 words, and 150 words or more). Here are the totals for each year in the four categories, from shortest to longest:

2012: 41, 67, 50, 43 (201 total)

2013: 68, 83, 51, 50 (252 total)

2014: 51, 54, 40, 41 (186 total)

2015: 145, 120, 56, 31 (352 total)

2016: 91, 134, 64, 77 (366 total)

As I already knew, the increased recall in 2015 happened at the shorter end of the word length spectrum.  My new approach to recall seemed to yield a lot of short dreams that I might not have remembered or recorded in previous years.  Then the 2016 dreams shifted again, with a more even distribution of word lengths, closer to the previous years but with higher total numbers.

Dividing the dreams into these subsets makes it possible to address the main question: what are the content differences between dreams of different lengths?

For each of the 20 subsets of dreams I used the SDDb 2.0 word search template to determine the frequencies for 40 categories of word usage, organized into 8 classes (Perceptions, Emotions, Characters, Cognitions, Social Interactions, Movement, Culture, and Elements).

The results of this analysis suggest that shorter dreams are not dramatically different from longer dreams in terms of the relative proportions of their word usage.  The raw percentages of word usage do rise from shorter to longer dreams, of course, but the relative proportions generally do not.

Consider the following excerpt from the analysis, which shows the four subsets of dreams from 2013, and the results of searching these reports for references to “Perception” words, from shortest to longest reports.  The numbers are percentages of the dreams that have at least one reference to the words in the category.

Vision: 24, 54, 69, 76

Hearing: 1, 6, 18, 22

Touch: 1, 6, 16, 26

Smell/Taste: 1, 2, 0, 6

Color: 21, 42, 31, 56

The longer dreams have more references to “Touch” than do the shorter dreams, but the longer dreams also have many more references to “Vision” and “Color” than to “Touch,” which is the same pattern found in the shorter dreams.  It’s this kind of pattern—the relative proportions between the various word categories—that remains consistent regardless of the length of the dreams.

This finding suggests the proportions among the word categories do not, for the most part, dramatically change across word lengths.  These proportions can be found in short, medium, and long dreams.  Even very short dreams preserve the basic architecture of typical dream content.

I need to do a more precise mathematical analysis of these patterns, to illuminate subtler variations that may alter my conclusions.  But I’m reassured by these initial results indicating that shorter dreams are just as legitimate as longer dreams for data-driven research and theorizing.

That’s the big picture.  Within this portrait of broad consistency, there are a few instances where the longer dreams do have an unusually high frequency of a particular word category.  The most prominent are Fear, Speech, Walking/Running, and Transportation.  These are the word categories that seem to be over-represented in longer dreams.  They are significant contributors to what makes long dreams so long.

Here is an example from the 2016 dreams to illustrate what I mean, using the Emotions class. The numbers are percentages of the dreams that have at least one reference to the words in the category, from shortest to longest reports.  Note the dramatic rise in Fear words across the four subsets.

Fear: 3, 11, 36, 55

Anger: 2, 4, 9, 17

Sadness: 4, 2, 8, 5

Wonder/Confusion: 23, 40, 63, 75

Happiness: 11, 22, 20, 26

The shortest dreams have scarcely any references to fear, whereas more than half the longest dreams have a reference to fear.

What I think this means is that when a dream introduces a reference to fear, it heightens my awareness of what’s going on in the dream space.  It stimulates an expansion of what I notice and find significant, and after awakening this requires a lengthier report to describe adequately.

What about the unusual increase in Speech references in longer dreams?  Perhaps a dream in which people start talking with each other is more likely to deepen the interaction and extend the overall experience.

Same with the increased references to Walking/Running and Transportation: a dream in which people are moving from one place to another is probably going to include additional details about what happens before, during, and after the movement.

So here’s a more refined conclusion: Shorter dreams are mostly similar to longer dreams in their basic content patterns, except that longer dreams tend to be scarier, more mobile, and more conversational.

Looking specifically at the 2015 dreams, I found the word usage frequencies were mostly lower compared to previous years, but they generally stayed the same in terms of their relative proportions to each other.  Even though the 2015 dreams were much shorter than the dreams of previous years, they shared with the other dreams a consistent profile of relative frequencies across all the word categories.  So my increased recall that year did not significantly alter the content patterns of the dreams.

Finally, I thought it would be fun to try a “blind analysis” of my own dreams.  Now that I have identified this remarkably stable profile of my dream content over five years of time, including both short dreams and long dreams, what do the patterns reveal about my life?

If I pretend that these dreams came from a stranger about whom I have no biographical knowledge, I would predict that in waking life this person:

Is male

Is visually oriented

Often experiences wonder/confusion

Is sexually active

Cares about his wife

Cares about cats

Has equal relations with men and women

Likes running

Is not concerned about death

Has lots of interactions with cars and streets

Likes basketball

Likes music and movies

Has lots of interactions with water and earth

All of these inferences are grounded in the statistical results of the word searches, and I would have to affirm every one of them as accurate.  Indeed, this is a remarkably concise summary of my concerns, interests, and activities in waking life.

Most importantly for the topic of this essay, the content patterns that helped me generate these inferences are observable in the shortest dreams.  I would have made most of these same accurate predictions if I had only been looking at the dreams of less than 50 words.

This means the answer to the opening question is no, there is not a significant difference in patterns of content between short and long dreams.  Perhaps dreams should be conceived as having a kind of fractal quality: even at a small scale they reflect the same basic structures that shapes things at a larger scale.

I will close by noting the three most striking discontinuities between the word usage frequencies in my dreams and the concerns, interests, and activities of my waking life.  These are instances where my blind analysis predictions would have been wrong.

First, I have very few references in my dreams to “Fantastic Beings,” which might lead to the inference that I do not like the cultural genres of science fiction or fantasy.  This is not true; I have always loved books, movies, and tv shows in the sci-fi and fantasy realm.  Perhaps what I like about these stories are not the odd characters (vampires, zombies, aliens, robots, etc.) but rather the spirit of unpredictable novelty and imaginative adventure.  Putting it in those terms, my high frequency of “Wonder/Confusion” words might be a better sign of my cultural interests in this direction.

Second, I have only moderate references to “Reading/Writing,” which might suggest I do not engage much with these activities.  This is not true; I am a voracious reader and prolific writer, and have been so for several decades.  What strikes me as discontinuous is that my dreams don’t have far more references to reading and writing, given their central importance in my waking life.  Ernest Hartmann’s notion that we typically do not dream of the three R’s might be a factor here.

And third, I have very few references to “Religion,” which would prompt the inference that I have little or no concern about religion.  At one level this is definitely false; I have a Ph.D. in religious studies and I read and write about religion very frequently.  One would never know this about my waking life based only on the patterns of my dreams.  And yet, at another level this inference is surely true; I was not raised in a religious household, I do not personally identify with any official religious tradition, and I rarely attend religious worship services.  Perhaps this all makes sense in that religion is an important intellectual category for me, but it is not a personal concern.  My spiritual pursuits are more likely to be expressed in dreams with references to other word categories like water, art, sexuality, animals, and flying.

That’s as far as I’ve gotten.  The next step will be trying this same process of analysis with other sets and series of dreams.


Note: this post was originally published in Psychology Today on May 4, 2017.