DreamResearch.net: The Case for a Cognitive Theory of Dreams

Update, July 2022

This unpublished working paper from 2010 was the foundation of Bill Domhoff's 2022 book, The Neurocognitive Theory of Dreaming. For a summary of the ideas in the book, a 12-page PDF handout is available.

Introduction and Overview

Four very different types of unexpected research findings from inside and outside the sleep laboratory since the 1950s make it possible to suggest a new cognitive approach to dreaming and dream content, an approach that has the potential to be extended into a neurocognitive theory as well. These findings, which are discussed throughout this article, cast doubt on the Freudian, Jungian, and activation-synthesis theories that dominated thinking about dreams in the twentieth century. Those three theories all began with the idea that there were major differences between waking cognition and dreaming, but the findings presented in this article suggest that there are far more parallels between dreaming and waking thought than they realized (Domhoff, 2003b).

If there are more similarities than differences between dreaming and waking cognition, then there may be only small changes when alert waking thought turns into dreaming. Therefore, the key issue is not the activation of a "primary process" and "repressed" wishes, as in Freudian theory, nor the expression of "archetypal symbols" lodged within an inherited "collective unconscious," as in Jungian theory. Nor is there a need for allegedly random stimulation from the pontine tegmentum within the brain stem or for the unique neurochemical state that is part of REM sleep, as in activation-synthesis theory (Hobson, 2002; Hobson, Pace-Schott, & Stickgold, 2000b).

Instead, dreaming can be seen as the "default" position for the activated brain when it is not forced to focus on physical and social reality by (1) external stimuli and (2) the self system that reminds us of who we are, where we are, and what the tasks are that face us. This possibility is supported by laboratory studies revealing the dreamlike nature of thinking during the transition from waking to sleeping (Foulkes & Vogel, 1965; Vogel, 1991; Vogel, Barrowclough, & Giesler, 1972). It is also supported by systematic observations demonstrating that dreaming can occur during relaxed waking states when lying quietly in a darkened room, with wakefulness monitored by the EEG (Foulkes, 1999; Foulkes & Fleisher, 1975; Foulkes & Scott, 1973). There are also strong cognitive connections between waking fantasy, daydreaming, and dreaming (Levin & Young, 2001-2002). Furthermore, these findings on similarities between dreaming and daydreaming may provide a bridge to studies of mind-wandering and the default network that seems to underlie drifting waking thought (e.g., Mason, et al., 2007; Smallwood & Schooler, 2006).

Based on the evidence for parallels between dreaming and waking cognition, what makes a cognitive approach to dreams distinctive is that it begins with concepts derived from laboratory studies of waking thought. The emphasis is on the fact that thinking and imagining develop as part of a "conceptual system," that is, a system of schemas and scripts, which is the organizational basis for all human knowledge and beliefs. Drawing on the same conceptual system that underlies waking thought, dreaming occurs whenever there is (1) an intact and fully mature neural network for dreaming; (2) an adequate level of cortical activation; (3) an occlusion of external stimuli; and (4) the loss of conscious self-control, i.e., a shutting down to the cognitive system of "self" (Foulkes, 1999)

From a cognitive perspective, dreams express people's "conceptions," which are also the basis for action in the waking world. Dreams are a dramatic and perceptible embodiment of schemas, scripts, and general knowledge. They are like plays that the mind stages for itself when it doesn't have anything specific to do. In particular, many dream scenarios express several key aspects of people's conceptual systems, especially self-conceptions, which can be defined as a set of cognitive generalizations about the person that guide the processing of self-relevant information and events. Starting with the idea that dreams often reveal highly personal conceptions, it is possible to build a complex picture of a dreamer's overall conceptual system because people usually have more than one conception of themselves. Dreams also express conceptions of family and friends; they less often involve politics, economics, or other current events. However, there is also a significant minority of dreams, perhaps as many as 30% for some adults, which have no easily discernable connections to the person's waking life (Domhoff, Meyer-Gomes, & Schredl, 2005-2006). They are more like sagas or adventure stories; Foulkes (1999, p. 136) calls such dreams "narrative-driven" to contrast them with dreams that seem to be based on personal concerns. (For more detail on these issues, read my 2003 article about a cognitive theory of dreams; specifically, the section entitled "Dream Content and Waking Cognition.")

A cognitive approach also contains a way to assess the weight to be given to the conceptions expressed in dreams: by determining the relative frequency of their occurrence. Numerous studies show that the frequency with which a person, action, or activity occurs in a series of dreams reveals the "intensity" of the "concern" with that person, action, or activity in waking life, which means that dreams are dramatized enactments of both "conceptions" and "concerns" (Domhoff, 1996, 2003b). The emphasis on "concerns" links dreaming with the fact that the drift of waking thought is shaped to a great extent by underlying concerns (Klinger, 1978, 1999)

In the following sections I turn to the evidence that led to this theory. It comes from (1) laboratory studies of the Rapid Eye Movement (REM) and non-REM (NREM) dream reports of adults; (2) laboratory investigations of the development of dreaming in children; (3) studies of large samples of dreams collected outside the laboratory from groups and individuals; and (4) neuropsychological studies of patients who experienced changes in their dreaming as a by-product of brain lesions.

Before I do so, however, it might be useful to avoid any potential confusion by mentioning the four interrelated meanings of the word "dream" that follow one from the other. First, a "dream" is a form of thinking during sleep that, as already briefly stated, occurs when there is (1) an intact and fully mature neural network for dreaming; (2) an adequate level of cortical activation; (3) an occlusion of external stimuli by the sensory gates located in the thalamus; and (4) the loss of conscious self-control, i.e., a shutting down to the cognitive system of "self." Second, the word "dream" is used to describe a cognitive state that people "experience" as an ongoing narrative because the thought patterns simulate waking reality. Third, a "dream" is also what people remember in the morning, so it is in this sense a "memory" of the dreaming experience. Finally, a "dream" is the spoken or written "report" provided to researchers based on the memory of the dreaming experience. Put another way, there is a dreaming process, an experienced dream, a remembered dream, and a reported dream.

It also might be useful to mention the three ways in which dream reports differ from other types of oral utterances or written reports. First, dream reports are not generally self-initiated; in other words, very few of the people who provide dream reports would have written down or spoken their dreams if they had not been asked to do so by researchers, although those few who keep dream diaries for their own personal reasons are an important exception. Second, the dream reports that are used in dream research are a unique human statement in that they were not spoken or written -- even by those who keep personal dream journals -- in order to communicate with or influence other people. That is, they are "representational," not "instrumental," communications (Hall & Van de Castle, 1966, p. 21). Third, it is noteworthy that dreams are usually experienced as something that happens to the dreamer without self-effort, and therefore not as something intended, so people do not tend to accept as much responsibility for their dream reports as they do for what they say or write based on waking thoughts and experiences.

Laboratory Studies of Adults

The many laboratory studies of adult dreams in the 1960s and 1970s, now often forgotten or ignored, led to a very surprising result: the dreams reported from REM awakenings are usually reasonable simulations of the waking world that deal with everyday topics and contain relatively few fantastic or bizarre aspects; then, too, the speech acts included in dreams are as well executed and context-appropriate as in waking life (Foulkes, et al., 1993; Meier, 1993)

The most comprehensive study of REM dream content in the sleep laboratory was based on 635 dream reports collected "for a variety of experimental purposes" in a series of investigations over a period of seven years between 1960 and 1967 (Snyder, 1970, p. 127). The 58 young adult men and women who participated in these studies were awakened on 250 nights in two different laboratories. Although there were some small differences due to word length, the overall finding was that "dreaming consciousness" is a "remarkably faithful replica of waking life;" a prototypical REM dream report is a "clear, coherent, and detailed account of a realistic situation involving the dreamer and other people caught up in very ordinary activities and preoccupations, and usually talking about them"(Snyder, 1970, p. 133, 148). Overall, as many as 90% of the dream reports "would have been considered credible descriptions of everyday experience" (Snyder, Karacan, Tharp, & Scott, 1968, p. 375).

This conclusion is based primarily on a series of ratings for coherence, dramatic quality, credibility, and bizarreness. Sixty to 80% of the reports were highly coherent on a three point scale, as compared with less than 5% that were rated as low on coherence. Three-fourths had a "nil" or "low" degree of drama on a four-point scale, and less than 10% were high on drama. Fully 65% of the dream reports were rated as highly credible and another 25% as of medium credibility; about 8% were rated as low on credibility and 2% as having no credibility. In keeping with the findings on credibility, the dreams were rated as having a low degree of bizarreness. Fifty percent of the longest reports were rated as having no bizarreness, 30% as having a low degree of bizarreness, 8% as having a medium degree, and 2% as having a high degree (Snyder, 1970, pp. 145-146).

The apparent lack of highly unusual dream content in REM reports was investigated in more detail in a study of 16 young adult women who spent two consecutive nights each in the lab and answered questions about the familiarity and likelihood of specific dream elements after an average of four REM awakenings per night (Dorus, Dorus, & Rechtschaffen, 1971). The investigators concluded that their results "emphasize the rarity of the bizarre in dreams" because major distortions of actual waking experiences reach a high of only 16.7% of all the activities and social interactions, and of 6.2% and 7.8% for all characters and physical surroundings (Dorus, et al., 1971, p. 367). The figures for the most improbable category of events that were never experienced by the dreamer in waking life were 4.9% of all physical surroundings, 1.3% of all characters, and 6.8% of all activities and social interactions. When the investigators carried out global ratings of each dream for overall novelty, they found that 25.8% contained large but plausible differences from previous waking experiences and that 8.9% were highly improbable by waking standards.

In general, then, REM dreams are a more realistic enactment of everyday life than is suggested by the psychiatric tradition from which the Freudian, Jungian, and activation-synthesis theories derive. However, this does not mean there is no bizarreness in dreams, just far less than Freudian, Jungian, and activation-synthesis theorists assume. But the secondary nature of bizarreness in dreams also raises another question: what about bizarreness in waking thought? Studies of waking thought samples suggest that there are far more jumps in thinking and the sudden appearance of thoughts that seem to come out of nowhere than is implied by a juxtaposition of bizarre dreaming with rational waking thought (Kane, et al., 2007; Klinger, 1999; Klinger & Cox, 1987-1988).

Still other laboratory studies concluded that NREM dream reports do not differ very much from REM reports, especially late in the sleep period when participants are awakened from Stage II NREM (Antrobus, 1983; Antrobus, Kondo, & Reinsel, 1995; Cicogna, Natale, Occhionero, & Bosinelli, 1998). This conclusion was first denied and then downplayed by activation-synthesis theorists because of their emphasis on REM sleep as the basis for the allegedly bizarre nature of dreams (Hobson, et al., 2000b). More recently activation-synthesis theorists have found it more difficult to defend their REM-based theory of activation since results from one of their own studies demonstrated that REM and Stage II NREM reports are increasingly similar after the third REM period, as earlier studies also showed (R. Fosse, Stickgold, & Hobson, 2004). They now cling to the fact that two studies using their database of home-collected REM and NREM dream reports found differences in the frequency of aggression in the REM and Stage II NREM reports (McNamara, McLaren, & Durso, 2007; McNamara, McLaren, Smith, Brown, & Stickgold, 2005). In the first of these studies, 24% of 100 late-night REM reports had at least one hostility or aggression, compared to only 12% for the 100 NREM reports, but that leaves 82% of the dreams in the combined sample similar in that they contained no aggression. Moreover, there were no differences on friendly interactions in the first of the two studies. In the second study, there were more similarities than differences for reports from the two stages of sleep.

Taken together, the important point is that all these studies show that "real" dreaming can take place, at the least, in REM and Stage II NREM. The finding that the frequency of aggression varies in dreams from the two different states, which still needs to be replicated with another database before it is accepted as gospel, is an interesting subsidiary finding that fits with the fact that aggression is the element in dreams that differs the most by age, gender, culture, and method of dream collection (home vs. laboratory) (Avila-White, Schneider, & Domhoff, 1999; Domhoff, 1996; Domhoff & Schneider, 1999; Foulkes, 1979).

Thus, the activation-synthesis theorists are refuted in their emphasis on the brain stem origins of dreaming because slight differences in actual content, which activation-synthesis theorists usually dismiss as a secondary issue, are not enough to defend their weak claim that REM activation is the only basis for vivid dreaming. Nor does it make sense to resort to the claim that "covert REM sleep" is causing the Stage II NREM dreams (Hobson, 2000; Hobson, Pace-Schott, & Stickgold, 2000a; Nielsen, 2000). It is be more accurate to say that common sources are activating the brain in REM and Stage II NREM to a level where dreaming can occur (Wamsley, et al., 2007). This broader view fits with a cognitive theory of dreams, which emphasizes the level of brain activation, whatever its source.

Developmental Studies of Children

Longitudinal and cross-sectional studies of children ages 3 to 15 in the sleep laboratory reveal that dreaming gradually develops beginning late in the pre-school years in steps that parallel waking cognitive development (Foulkes, 1982, 1999; Foulkes, Hollifield, Sullivan, Bradley, & Terry, 1990). The median recall of pre-school children from REM awakenings of anything that could reasonably be called a dream was only 15%, and the content of the few reports that were obtained was static, undeveloped, and lacking in any emotion. The REM reports became more "dreamlike" (in terms of characters, themes, and actions) in the 5 to 6 year-olds, and the dreamers themselves began to play a more central role in their dreams by ages 7-8. Dreaming became adultlike in its cognitive complexity at ages 9-10, but it was not until the children were 11 to 13 years old that their dreams began to resemble those of adult laboratory participants in frequency, length, content, and emotions, or to have any relationship to personality.

Detailed studies of the waking cognitive abilities of the children in these studies showed that verbal and linguistic skills do not play a role until dreaming is fully developed. They also reveal that the lack of dream reports cannot be attributed to a failure to recall or an inability to communicate. The one good and consistent predictor of the frequency of dream reporting in children ages 5-9 is visuospatial skills, which leads to the hypothesis that mental imagery may develop gradually and be a necessary cognitive prerequisite for dreaming. This idea is supported by studies revealing that those who are born blind or who become blind before the age of 4 have no visual imagery in their dreams, whereas those who become blind after age 6 continue to have visual imagery in their dreams (N. Kerr, 1993).

Based on the results of these laboratory studies, dreaming appears to be a gradual cognitive achievement that depends upon the development of cognitive abilities that are also important in waking life. These results are compatible with a cognitive theory of dreams and call into question the claims about pre-school children's dreams by Freudians and Jungians, who rely on retrospective reports from adults or on home dream reports that are now suspect as possible confabulations due to pressures from adults to report a dream (Domhoff, 2003b, pp. 23-24). In addition, these findings were completely unanticipated by activation-synthesis theorists, who have refused to accept them. As if to underscore their scorn for a cognitive theory, the activation-synthesis theorists have stated that "we specifically suggest that the human neonate, spending as it does more than 50% of its time in REM sleep, is having indescribable but nevertheless real oneiric experiences" (Hobson, et al., 2000b, pp. 803). This statement shows that their theory begins and ends at the neurophysiological level, whereas a cognitive theory begins with phenomenal experience and dreams reports -- or the lack thereof. (For more detail, see the section entitled "The Development of Dreaming Cognition" in my 2003 article.)

Studies of Home-Reported Adult Dreams

Thanks to several studies demonstrating that dream reports collected in the laboratory are similar in most respects to those collected from home recall, except for the greater frequency of aggression in home dream reports (Domhoff, 2003b), it is possible for a cognitive approach to draw on studies of home dream reports using a rigorous and comprehensive coding system for dream content developed by Hall and Van de Castle (1966) to quantify home-reported dreams. This system of 10 general coding categories for settings, objects, characters, social interactions, activities, emotions, misfortunes, successes and failures, food and eating elements, and descriptive elements rests on the nominal level of measurement, uses percentages and ratios to correct for differences in dream length, and has good reliability (Domhoff, 2003b). Studies using this system reveal that the dream lives of college men and women in the United States remained about the same throughout the second half of the twentieth century. College students from several other industrialized democracies, including India and Japan, have similar patterns to the American students, as do teenagers and young adults in Argentina, Mexico, and Peru (Domhoff, 1996).

Despite the generally realistic nature of these home dream reports, the studies demonstrate that dreams are not a perfect simulation of everyday life, as also found in the laboratory studies. For example, in the Hall and Van de Castle (1966) normative samples of 500 male and 500 female dream reports, 7% of the familiar male settings and 14% of the familiar female settings were in some way different from the way they actually were in waking life, and almost 2% of the characters were dead or imaginary, or else turned into another character (that is, a metamorphosis occurred). It is also noteworthy that about one-third of all dream reports contained "misfortunes" that range from being lost to illness to the death of a loved one, and that the negative emotions of sadness, anger, confusion, and apprehension, when taken as a whole, greatly outnumber the expression of happiness.

In contrast to the changes in dream content from early childhood to adolescence that were uncovered in the sleep lab, dream content is extremely stable in terms of characters, social interactions, and most other dream elements after age 18 according to cross-sectional studies in the United States, Canada, and Switzerland, except for a possible decline in physical aggressions and negative emotions. Dream content thus seems to parallel the stability of adult personality (Domhoff, 1996, Chapter 5).

Within the context of these many well-established group findings, blind analyses of dream journals kept by individuals for personal, intellectual, or artistic reasons are useful as nonreactive measures not influenced by the purposes of the investigators who code and quantify them. Conclusions drawn from such nonreactive archival data are considered most reliable and useful when they are based on a diversity of archives likely to have different sources of potential biases. Studies of dream journals reveal an individual consistency in dream content that stretches from the late teens to old age, which is further support for a cognitive theory of dreams because of waking evidence that people's personalities change very little once they are adults (McCrae & Costa, 2003). People's dream lives vary from day to day and week to week, but consistency in both themes and quantitative content categories manifests itself through comparisons of hundreds of dream reports and with time spans of months and years (Domhoff, 1996, 2003b).

Blind analyses of dream journals also have made it possible to do detailed studies of specific aspects of dream content and their relationship to waking thought through the formulation of inferences that can be accepted or rejected by the dreamer and other respondents. Such studies lead to the conclusion that much, but not all, dream content is continuous with the dreamer's waking concerns and interests. The most direct continuities involve the main people in a dreamer's life and the nature of the social interactions with them. There also is considerable continuity for many of the dreamer's primary interests and activities. This continuity is not with day-to-day events, but with general concerns, which provides yet another parallel between dreaming and waking cognition. However, there are as many as 30% of dreams that do not seem related to waking concerns (Domhoff, 2003a; Domhoff, et al., 2005-2006).

The blind analyses based on inferences drawn from Hall/Van de Castle content analyses have been supplemented and independently verified by studies using strings of words entered into the search engine on dreambank.net, a large archive of over 23,000 dream reports that also includes several technologies for retrieving and studying specific elements of dream content (Schneider & Domhoff, 1999). In one such study it was shown that correct inferences could be made by a researcher who only knew the frequencies for 40 word strings; the inference concerned the dreamer's waking personality (e.g., socially outgoing, even tempered), the activities in which he is involved (e.g., works with newspapers, sexually active), cultural preferences (e.g., a science-fiction fan), and relationships with family members (Bulkeley & Domhoff, 2010). Another study of this type showed that the inferences drawn from word-string results paralleled and supplemented those based on Hall/Van de Castle codings (Bulkeley, 2009),

In concluding this overview of findings on dream content inside and outside the sleep laboratory, it should be noted that the coherence, consistency, and continuity of most dream content revealed in virtually every study is not what would be expected on the basis of Freudian, Jungian, or activation-synthesis theory. In fact, these studies contradict every specific hypothesis that was put forth by Freud in The Interpretation of Dreams (1900), as shown in detailed critiques of the theory and its current main proponent, Mark Solms (Domhoff, 2003b, pp. 136-143; 2004). In the case of Jungian theory, the finding of continuity between dreams and waking concerns is in direct contradiction to its claim that dreaming has a "compensatory" function (according to this key Jungian tenet, dreams purportedly reveal the underdeveloped aspects of the psyche that are not expressed in waking life) (Jung, 1974). As for activation-synthesis theory, it is most severely challenged at the content level by the realistic and coherent nature of dream content revealed in both laboratory studies and studies of lengthy dream journals that stretch over decades.

Making a Cognitive Approach Neurocognitive

Asking neurological patients if their injuries had caused any changes in their dreaming led to the unanticipated discovery that different types of deficits and excesses of dreaming have waking cognitive parallels. Such a finding is further support for a cognitive approach to dreams because it suggests that dreaming and imaginative waking cognition may be dependent upon the same brain networks. For example, the reported loss of the ability to produce visual dream imagery in some patients studied in the sleep lab is paralleled by their reported loss of waking visual imagery (N. Kerr, 1993; N. Kerr & Foulkes, 1981; N. H. Kerr, Foulkes, & Jurkovic, 1978).

Building on this and many other examples of changes in both waking visual imagery and dream imagery, it might be possible to extend a cognitive approach in a neurophysiological direction and make it neurocognitive. This neurocognitive model would start with neuropsychological work from the late 1980s relating certain brain lesions, as determined by CAT scans, to the alteration or elimination of dreaming (Solms, 1997). It also would draw on brain-imaging studies comparing wakefulness, REM, and NREM (Braun, et al., 1997; Braun, et al., 1998; Maquet, 2000; Maquet, et al., 1996; Nofzinger, Mintun, Wiseman, Kupfer, & Moore, 1997). Both types of studies suggest roughly the same neural network as the basis for the conceptual processes involved in dreaming (Domhoff, 2003b, Chapter 1).

This neural network encompasses the limbic, paralimbic, and association cortices, with little or no role for the dorsolateral prefrontal cortex, sensorimotor cortex, or primary visual cortex. Thus, the association cortices, paralimbic structures, and limbic structures may operate as a closed loop to generate the process of dreaming. On the one hand, this subsystem is cut off from the primary sensory cortices that provide information about the external world, and on the other from the prefrontal cortices that integrate incoming sensory information with memory and emotion in the process of decision-making. This model implies that an unconstrained and freewheeling conceptual system can operate when there is sufficient brain activation. At the same time, it is clear from the nature of dream content that the neural network for dreaming contains enough cognitive processing areas, such as the medial frontal cortex and anterior cingulate cortex, to produce coherent dramatizations that often reflect the dreamer's conceptions and concerns in waking life. (For a detailed account that brings together the early neuropsychological and brain imaging studies, go to my 2003 summary, which fits with more recent findings except on a few details.)

Do Dreams Have an Adaptive Function?

There are many different theories concerning the possible adaptive function of dreaming. The most famous and intriguing theory of dream function, Freud's (1900, p. 180) idea that dreams are the "guardians of sleep," that is, a response to bodily-based wishes, seems to be refuted by several different types of findings. First, the frequency and regularity of dreaming in most people suggests that the process cannot be primarily a way to deal with wishes that emerge episodically during sleep, as Freud claimed. Second, there is very little dream content that seems to relate to the wishes Freud had in mind -- for food, drink, and sex. Third, there is every reason to believe that pre-school children seldom or never dream, but they sleep very soundly nonetheless (Foulkes, 1982). Fourth, lobotomized schizophrenics showed normal sleep in the laboratory, but they rarely reported dreams from REM awakenings (Jus, et al., 1973). Fifth, there are neurological patients who lose dreaming but retain the ability to sleep (Solms, 1997). Finally, there are even a few normal adults who do not dream, as shown through REM awakenings over two or three nights in the sleep laboratory (Pagel, 2003).

Nor, as noted earlier, at the end of the discussion of dream content, is there any support for Jung's well-known idea that most dreams, and especially those with roots in the "collective unconscious," have a compensatory function. Although this idea is very difficult to refute in a definitive way because there may be subtle forms of compensation, every relevant systematic study suggests that most dream content is continuous with waking thought or personality rather than compensatory (Domhoff, 2003b, pp. 144-147 for a critique of Jungian theory).

Activation-synthesis theorists argue that dreaming is the by-product of the "off-line" cognitive processing and sleep-dependent memory consolidation that allegedly occur during all stages of sleep, with some stages of sleep enhancing one kind of learning and other stages aiding other kinds (Stickgold, 2005). There are several major problems with this claim. First, there are serious methodological problems with the research on memory consolidation during sleep that have not been answered; it seems unlikely to some sleep researchers that very many, if any, of the claims about any special role for sleep in enhancing memory consolidation will be sustained (J. Siegel, 2001; J Siegel, 2005; Vertes & Eastman, 2000; Vertes & Siegel, 2005).

Second, the best case for memory consolidation during sleep concerns procedural memory, such as learning to weave a basket, which would not seem to have much to do with dreaming. Third, if dreaming is somehow intertwined with the consolidation of memories from the previous day, then it seems surprising that there are very few episodic memories in dreams (Baylor & Cavallero, 2001; M. Fosse, Fosse, Hobson, & Stickgold, 2003) and that only half of dreams at best have even one element that relates to events of the previous day (Botman & Crovitz, 1989; Harlow & Roll, 1992; Hartmann, 1968; Marquardt, Bonato, & Hoffmann, 1996; Nielsen & Powell, 1992). Fourth, if the function of dreaming is to process the emotional content of recent events in a person's life, as Stickgold (2005) claims, then the lack of emotion in the dreams of children under age 12 needs to be explained (Foulkes, 1982; Foulkes, et al., 1990), as does the absence of emotion in as many as 25-30% of dream reports from REM periods in three different studies (Domhoff, 2003b). Given these many problems, it is unlikely that any memory processing that might occur in sleep would have anything to do with dreaming except for its possible involvement in creating the level of activation necessary for dreaming to occur. Thus, it is more likely that activation-synthesis theorists will prove to be as wrong on this issue as they were on their claims for the location of the REM activation system (Sastre, Sakai, & Jouvet, 1981; J. Siegel & McGinty, 1977; J. Siegel, McGinty, & Breedlove, 1977), the nature of the interaction between REM and NREM sleep (Jones, 2005), and the role of brain stem signals in creating allegedly bizarre dream content (Jones, 2000; Pivik, 1991).

Several clinically oriented dream researchers from outside the Freudian, Jungian, and activation-synthesis realms believe that dreams have a general problem-solving function. However, this idea seems unlikely when it is realized that most people remember only a few percent of their many nightly dreams. More importantly, dreams collected inside and outside the sleep laboratory rarely have even the hint of a solution to a problem, and most of the anecdotal examples that are provided by proponents of the theory actually involve reveries, drug-induced states, or thoughts while falling asleep or waking up (Domhoff, 2003b, pp. 158-162).

Facing up to the rarity of dream recall and the lack of solutions to problems in most of the dreams that are recalled, some advocates of problem-solving theories now claim that only dramatic emotional dreams have a problem-solving function. There are two important distinctions that have to be made in considering this possibility. First, it is one thing for a dream to "reflect" a problem, as many dreams surely do; it is quite another for the dream content to offer a "solution." Second, a distinction has to be made between solutions that are present within a dream and waking ideas that are based upon thinking about the dream. New insights in the waking state are a much more plausible alternative because there is evidence that conscious attention is usually needed for problem solving (Blagrove, 1992, 1996, 2000).

As the foregoing discussion of rival theories of dream function suggests, there are many findings that contradict the idea that dreams have any kind of physiological or psychological function. It therefore makes sense to consider the possibility that dreams do not have an adaptive function. This hypothesis is consistent with a cognitive theory of dreams, which views dreaming as a complex cognitive achievement that is dependent upon the evolution of cognitive skills that may only be present in humans after they reach the age of 5 or 6 (Domhoff, 2003b, pp. 163-164). While the likely lack of dreaming in other mammals and pre-school children does not preclude an adaptive function for dreams in human adults, it does make it less probable, and it narrows the focus as to what that function might be if we recall that less than 5% of dreams are remembered by most adults and that virtually none of those dreams can be claimed to provide new insights or the solution to problems..

It therefore seems plausible that dreaming may not have any adaptive function. Just as the human love of and aptitude for music may be "an accidental by-product of traits that evolved for other purposes," (McDermott, 2009, p.164), so too the ability to dream may be a by-product of at least three specific cognitive abilities that have great adaptive value in the waking world: the ability to generate mental imagery, the ability to organize our experience in narrative form, and the ability to create an autobiographical self. It is noteworthy that all three of these abilities seem to develop between the ages of 3 and 6, which may be why there seems to be little or no dreaming before age 6. From that point on dreams occur, as already noted, simply because there is sufficient brain activation in a context where there is little or no guidance for the brain from external stimuli or the self-system.

Because dreams often seem to draw upon the same conceptions and concerns that are central to a person's waking life, they do have psychological meaning, and they therefore can portray central emotional preoccupations in creative and dramatic ways, but this does not necessarily mean that they have a purpose. From a cognitive point of view, function and meaning can be distinguished, a possibility that is rarely raised within the context of other dream theories.

Moreover, due to the fact that dreams have psychological meaning, people in many different cultures have developed "uses" for them in the course of history, including creative inspiration and personal exploration in our own culture. But emergent cultural uses, as important as they can be in human lives, are not the same as evolved psychological functions; not everything with a form or structure has an adaptive function (Thompson, 2000). In summary, and in a phrase that could be a mantra for a cognitive theory of dreams, dreams have psychological meaning and cultural uses, but it is unlikely at this point that they have any adaptive function.

Future Directions

Several different kinds of research projects could be carried out to test hypotheses generated by a cognitive theory of dreams. First, inquiries could be focused on pinpointing the small changes in cognitive functioning that might be responsible for the shift from relaxed waking thoughts, such as daydreaming and reverie, to the dramatic, self-involved, sensory-laden narratives called dreams. Studies of the imagery at sleep onset might be helpful in this regard (Nielsen, 1995), as might further studies of the relationship between daydreaming and dreaming (Levin & Young, 2001-2002).

Second, hypotheses related to a cognitive theory of dreams could be tested by in-depth examinations of both non-dreamers and those who recall several dreams a night in great detail. Comparisons of these two small "extreme" groups on cognitive and neuropsychological tests might prove especially useful.

Third, in-depth neuropsychological studies are needed of those people with brain lesions in the medial prefrontal cortex, the anterior cingulate cortex, and the basal forebrain who also suffer from excessive vivid dreaming during the night and the intrusion of dreaming into waking life. They are of great potential interest for a cognitive theory because some of them say that their waking thoughts quickly turn into pictures or that they feel like they are always dreaming, a claim that is supported by hospital staff members who report that these patients suffer from a confusion between dreaming and waking thought (Solms, 1997, pp. 198-199). So, too, Damasio et. al. (1985, p. 269) describe their patients with medial prefrontal lesions as suffering from "waking dreams;" Whitty and Lewin (1957) report several similar cases. It is also noteworthy that the medial prefrontal cortex is involved in processes of arousal and attention in waking life, and that injuries to this area can lead to confabulation and compulsive fabrication.

Fourth, in-depth investigations of dream journals from a few excellent recallers might help to explain the aspects of dream content that are not continuous with waking conceptions and concerns. These anomalous aspects of dream content may be the products of metaphoric thinking, although very little progress has been made in testing this cognitively based hypothesis (Domhoff, 2003b). Or it may be that unusual juxtapositions, blended settings, metamorphoses, and sudden scene changes reveal the limits of the mind under the conditions that produce dreaming (Domhoff, 2007; Foulkes, 1999).

Finally, there is a need for studies that compare dreams and waking thought samples from the same participants. Based on my reading of the protocols, the one effort along this line was inadequate because most of the participants simply reported what they were doing when they responded to a pager during the day (e.g., they report that they were eating lunch with friend, or walking to class, or sitting in class, or studying) (R. Fosse, Stickgold, & Hobson, 2001; Stickgold, Malia, Fosse, & Hobson, 2001). Contrary to the findings from that flawed database, other studies focused strictly on waking cognition provide evidence of frequent thought intrusions, mind wandering, attention lapses, and daydreaming, along with wide individual differences that make it essential to have the same participants in both conditions in order to say anything meaningful about the bizarreness of dreams in general (Kane, et al., 2007; Klinger, 1999; Smallwood & Schooler, 2006). It also would be useful to see if the default neural network in which mind wandering most often occurs has any relationship to the neural network for dreaming (Mason, et al., 2007), and if the areas involved in waking mental simulation have the degree of overlap with the neural network for dreaming that I suspect they do (Szpunar, 2010, for a summary of brain imaging studies of simulation tasks).

Conclusion

The systematic research findings briefly overviewed in this article add up to a strong refutation of the three traditional dream theories that predominated in the twentieth century. At the same time, these findings are consistent with a cognitive theory of dreams. This does not mean that the cognitive theory is now firmly established. But it does mean that such a theory is deserving of further testing. Indeed, it is the best current possibility for understanding the ongoing puzzle of the origin and meaning of dreams because using the most plausible ideas to explain the most systematic data is the hallmark of a scientific approach. By that standard, a cognitive theory of dreams is the only current theory that can encompass two well-established findings (1) dream content is for the most part coherent, consistent, and continuous with waking concerns; and (2) dreaming is a gradually developed cognitive achievement based on the ability to imagine and to generate mental imagery. Dreams are a perceptual embodiment and dramatization of a person's thoughts, an fMRI of the mind at night. They are the quintessential cognitive simulation because they are experienced as real while they are happening.

References

Antrobus, J. (1983). REM and NREM sleep reports: Comparisons of word frequencies by cognitive classes. Psychophysiology, 20, 562-568.

Antrobus, J., Kondo, T., & Reinsel, R. (1995). Dreaming in the late morning: Summation of REM and diurnal cortical activation. Consciousness & Cognition, 4, 275-299.

Avila-White, D., Schneider, A., & Domhoff, G. W. (1999). The most recent dreams of 12-13 year-old boys and girls: A methodological contribution to the study of dream content in teenagers. Dreaming, 9, 163-171.

Baylor, G., & Cavallero, C. (2001). Memory sources associated with REM and NREM dream reports throughout the night: A new look at the data. Sleep, 24, 165-170.

Blagrove, M. (1992). Dreams as a reflection of our waking concerns and abilities: A critique of the problem-solving paradigm in dream research. Dreaming, 2, 205-220.

Blagrove, M. (1996). Problems with the cognitive psychological modeling of dreaming. Journal of Mind and Behavior, 17, 99-134.

Blagrove, M. (2000). Dreams have meaning but no function. Behavioral and Brain Sciences, 23, 910.

Botman, H., & Crovitz (1989). Dream reports and autobiographical memory. Imagination, Cognition and Personality, 9, 213-214.

Braun, A., Balkin, T., Wesensten, N., Carson, R., Varga, M., Baldwin, P., et al. (1997). Regional cerebral blood flow throughout the sleep-wake cycle: An (H2O)-O-15 PET study. Brain, 120, 1173-1197.

Braun, A., Balkin, T., Wesensten, N., Gwadry, F., Carson, R., Varga, M., et al. (1998). Dissociated pattern of activity in visual cortices and their projections during human rapid eye movement sleep. Science, 279, 91-95.

Bulkeley, K. (2009). Seeking patterns in dream content: A systematic approach to word searches Consciousness and Cognition, 18, 909-916.

Bulkeley, K., & Domhoff, G. W. (2010). Detecting meaning in dream reports: An extension of a word search approach. Dreaming, In Press.

Cicogna, P., Natale, V., Occhionero, M., & Bosinelli, M. (1998). A comparison of mental activity during sleep onset and morning awakening. Sleep, 21(5), 462-470.

Damasio, A., Graff-Radford, N., Eslinger, P., Damasio, H., & Kassell, N. (1985). Amnesia following basal forebrain lesions. Archives of Neurology, 42, 263-271.

Domhoff, G. W. (1996). Finding meaning in dreams: A quantitative approach. New York: Plenum.

Domhoff, G. W. (2003a). Dreaming - An introduction to the science of sleep. Science, 299(5615), 1987-1988.

Domhoff, G. W. (2003b). The scientific study of dreams: Neural networks, cognitive development, and content analysis. Washington, DC: American Psychological Association.

Domhoff, G. W. (2004). Why did empirical dream researchers reject Freud? A critique of historical claims by Mark Solms. Dreaming, 14, 3-17.

Domhoff, G. W. (2007). Realistic simulation and bizarreness in dream content: Past findings and suggestions for future research. In D. Barrett & P. McNamara (Eds.), The new science of dreaming: Content, recall, and personality correlates (Vol. 2, pp. 1-27). Westport, CT: Praeger.

Domhoff, G. W., Meyer-Gomes, K., & Schredl, M. (2005-2006). Dreams as the expression of conceptions and concerns: A comparison of German and American college students. Imagination, Cognition & Personality, 25, 269-282.

Domhoff, G. W., & Schneider, A. (1999). Much ado about very little: The small effect sizes when home and laboratory collected dreams are compared. Dreaming, 9, 139-151.

Dorus, E., Dorus, W., & Rechtschaffen, A. (1971). The incidence of novelty in dreams. Archives of General Psychiatry, 25, 364-368.

Fosse, M., Fosse, R., Hobson, J. A., & Stickgold, R. (2003). Dreaming and episodic memory: A functional dissociation? Journal of Cognitive Neuroscience, 15, 1-9.

Fosse, R., Stickgold, R., & Hobson, J. A. (2001). Brain-mind states: Reciprocal variation in thoughts and hallucinations. Psychological Science, 12, 30-36.

Fosse, R., Stickgold, R., & Hobson, J. A. (2004). Thinking and hallucinating: Reciprocal changes in sleep. Psychophysiology, 41, 298-305.

Foulkes, D. (1979). Home and laboratory dreams: Four empirical studies and a conceptual reevaluation. Sleep, 2, 233-251.

Foulkes, D. (1999). Children's dreaming and the development of consciousness. Cambridge, MA: Harvard University Press.

Foulkes, D., & Fleisher, S. (1975). Mental activity in relaxed wakefulness. Journal of Abnormal Psychology, 84, 66-75.

Foulkes, D., Hollifield, M., Sullivan, B., Bradley, L., & Terry, R. (1990). REM dreaming and cognitive skills at ages 5-8: A cross-sectional study. International Journal of Behavioral Development, 13, 447-465.

Foulkes, D., Meier, B., Strauch, I., Kerr, N., Bradley, L., & Hollifield, M. (1993). Linguistic phenomena and language selection in the REM dreams of German-English bilinguals. International Journal of Psychology, 28(6), 871-891.

Foulkes, D., & Scott, E. (1973). An above-zero baseline for the incidence of momentarily hallucinatory mentation. Sleep Research, 2, 108.

Foulkes, D., & Vogel, G. (1965). Mental activity at sleep onset. Journal of Abnormal Psychology, 70, 231-243.

Freud, S. (1900). The interpretation of dreams (J. Crick, Trans.). London: Oxford University Press.

Hall, C., & Van de Castle, R. (1966). The content analysis of dreams. New York: Appleton-Century-Crofts.

Harlow, J., & Roll, S. (1992). Frequency of day residue in dreams of young adults. Perceptual & Motor Skills, 74, 832-834.

Hartmann, E. (1968). The day residue: Time distribution of waking events. Psychophysiology, 5, 222.

Hobson, J. A. (2000). The ghost of Sigmund Freud haunts Mark Solms's dream theory. Behavioral and Brain Sciences, 23, 951-952.

Hobson, J. A. (2002). Dreaming: An introduction to the science of sleep. New York: Oxford University Press.

Hobson, J. A., Pace-Schott, E. F., & Stickgold, R. (2000a). Dream science 2000: A response to commentaries on dreaming and the brain. Behavioral and Brain Sciences, 23, 1019-1034.

Hobson, J. A., Pace-Schott, E. F., & Stickgold, R. (2000b). Dreaming and the brain: Toward a cognitive neuroscience of conscious states. Behavioral and Brain Sciences, 23(6), 793-842.

Jones, B. E. (2000). The interpretation of physiology. Behavioral and Brain Sciences, 23, 955-956.

Jones, B. E. (2005). Basic mechanisms of sleep-wake states. In M. Kryger, T. Roth & W. Dement (Eds.), Principles and practices of sleep medicine (4th ed., pp. 136-153). Philadelphia: Elsevier Saunders.

Jus, A., Jus, K., Villenueve, A., Pires, A., Lachance, R., Fortier, J., et al. (1973). Studies on dream recall in chronic schizophrenic patients after prefrontal lobotomy. Biological Psychiatry, 6, 275-293.

Kane, M., Brown, L., McVay, J., Silvia, P., Myin-Germeys, I., & Kwapil, T. (2007). For whom the mind wanders, and when: An experience-sampling study of working memory and executive control in daily life. Psychological Science, 18, 614-621.

Kerr, N. (1993). Mental imagery, dreams, and perception. In D. Foulkes & C. Cavallero (Eds.), Dreaming as Cognition (pp. 18-37). New York: Harvester Wheatsheaf.

Kerr, N., & Foulkes, D. (1981). Right hemispheric mediation of dream visualization: A case study. Cortex, 17, 603-610.

Kerr, N. H., Foulkes, D., & Jurkovic, G. J. (1978). Reported absence of visual dream imagery in a normally sighted subject with Turner's syndrome. Journal of Mental Imagery, 2, 247-264.

Klinger, E. (1978). Modes of normal conscious thought flow. In K. S. Pope & J. L. Singer (Eds.), The stream of consciousness (pp. 225-258). New York: Plenum Press.

Klinger, E. (1999). Thought flow: Properties and mechanisms underlying shifts in content. In J. Singer & P. Salovey (Eds.), At play in the fields of consciousness (pp. 29-50). Hillsdale, NJ: Erlbaum.

Klinger, E., & Cox, W. (1987-1988). Dimensions of thought flow in everyday life. Imagination, Cognition, and Personality, 7, 105-128.

Levin, R., & Young, H. (2001-2002). The relation of waking fantasy to dreaming. Imagination, Cognition & Personality, 21, 201-219.

Maquet, P. (2000). Functional neuroimaging of normal human sleep by positron emission tomography. Journal of Sleep Research, 9, 207-231.

Maquet, P., Peters, J.-M., Aerts, J., Delfiore, G., Dequerldre, C., Luxen, A., et al. (1996, September 12). Functional neuroanatomy of human rapid-eye-movement sleep and dreaming. Nature, 383, 163-166.

Marquardt, C. J. G., Bonato, R. A., & Hoffmann, R. F. (1996). An empirical investigation into the day-residue and dream-lag effects. Dreaming, 6, 57-65.

Mason, M., Norton, M., Van Horn, J., Wnger, D., Grafton, S., & Macrae, N. (2007). Wandering minds: The default network and stimulus-independent thought. Science, 315, 393-395.

McCrae, R., & Costa, P. (2003). Personality in adulthood: A five-factor theory perspective (2nd ed.). New York: Guilford Press.

McDermott, J. H. (2009). What can experiments reveal about the origins of music? Current Directions in Psychological Science, 18, 164-168.

McNamara, P., McLaren, D., & Durso, K. (2007). Representation of the self in REM and NREM dreams. Dreaming, 17 113-126.

McNamara, P., McLaren, D., Smith, D., Brown, A., & Stickgold, R. (2005). A 'Jekyll and Hyde' within: Aggressive versus friendly interactions in REM and NREM dreams. Psychological Science, 16, 130-136.

Meier, B. (1993). Speech and thinking in dreams. In C. Cavallero & D. Foulkes (Eds.), Dreaming as cognition (pp. 58-76). New York: Harvester Wheatsheaf.

Nielsen, T. (1995). Describing and modeling hypnagogic imagery using a systematic self-observation procedure. Dreaming, 5, 75-94.

Nielsen, T. (2000). A review of mentation in REM and NREM sleep: "Covert" REM sleep as a possible reconciliation of two opposing models. Behavioral and Brain Sciences, 23, 851-866.

Nielsen, T., & Powell, R. (1992). The day-residue and dream-lag effect. Dreaming, 2, 67-77.

Nofzinger, E., Mintun, M., Wiseman, M., Kupfer, D., & Moore, R. (1997). Forebrain activation in REM sleep: an FDG PET study. Brain Research, 770, 192-201.

Pivik, R. T. (1991). Tonic states and phasic events in relation to sleep mentation. In S. Ellman, J. Antrobus & . (Eds.), The mind in sleep: Psychology and psychophysiology (2nd ed., pp. 214-247). New York: Wiley.

Sastre, J.-P., Sakai, K., & Jouvet, M. (1981). Are the gigantocellular tegmental field neurons responsible for paraxodical sleep? Brain Research, 229(1), 147-161.

Siegel, J. (2001). The REM sleep-memory consolidation hypothesis. Science, 294, 1058-1063.

Siegel, J. (2005). Clues to the function of mammalian sleep. Nature, 437, 1264-1271.

Siegel, J., & McGinty, D. (1977). Pontine reticular formation neurons: Relationship of discharge to motor activity. Science, 196, 678-680.

Siegel, J., McGinty, D., & Breedlove, S. (1977). Sleep and waking activity of pontine gigantocellular field neurons. Experimental Neurology, 56, 553-573.

Smallwood, J., & Schooler, J. (2006). The restless mind. Psychological Bulletin 132, 946 -- 958.

Snyder, F. (1970). The phenomenology of dreaming. In L. Madow & L. Snow (Eds.), The psychodynamic implications of the physiological studies on dreams (pp. 124-151). Springfield, IL: Thomas.

Snyder, F., Karacan, I., Tharp, V., & Scott, J. (1968). Phenomenology of REM dreaming. Psychophysiology, 4, 375.

Solms, M. (1997). The neuropsychology of dreams: A clinico-anatomical study. Hillsdale, NJ: Erlbaum.

Stickgold, R. (2005). Why we dream. In M. Kryger, T. Roth & W. Dement (Eds.), Principles and practices of sleep medicine (4th ed., pp. 579-587). Philadelpha: Elsevier Saunders.

Stickgold, R., Malia, A., Fosse, R., & Hobson, J. A. (2001). Brain-mind states: Longitudinal field study of sleep/wake factors influencing mentation report length. Sleep, 24(2), 171-179.

Szpunar, K. (2010). Episodic future thought: An emerging concept. Perspectives on Psychological Science, 5, 142-162.

Thompson, N. (2000). Evolutionary psychology can ill afford adaptionist and mentalist credulity. Behavioral and Brain Sciences, 23, 1013-1014.

Vertes, R., & Eastman, K. (2000). The case against memory consolidation in REM sleep. Behavioral and Brain Sciences, 23, 867-876.

Vertes, R., & Siegel, J. (2005). Time for the sleep community to take a critical look at the purported role of sleep in memory processing. Sleep, 28, 1228-1229.

Vogel, G. (1991). Sleep-onset mentation. In S. Ellman & J. Antrobus (Eds.), The mind in sleep: Psychology and psychophysiology (2nd ed., pp. 125-136). New York: Wiley & Sons.

Vogel, G., Barrowclough, B., & Giesler, D. (1972). Limited discriminability of REM and sleep onset reports and its psychiatric implications. Archives of General Psychiatry, 26, 449-455.

Wamsley, E. J., Hirota, Y., Tucker, M. A., Smith, M. R., Doan, T., & Antrobus, J. S. (2007). Circadian and ultradian influences on dreaming: A dual rhythm model. Brain Research Bulletin, 71, 347-354.

Whitty, C., & Lewin, W. (1957). Vivid day-dreaming: An unusual form of confusion following anterior cingulectomy. Brain, 80, 72-76.

The Case for a Cognitive Theory of Dreams

G. William Domhoff

University of California, Santa Cruz