Dreaming -motivated or meaningless?
There are two widely held views about dreams. One view is that they contain information about our waking life, including our waking cognition and emotions, and possibly even some information that we are motivated to ignore when awake: On this view, dreams are worth examining. The other view is that dreaming is a delirium, dreams are senseless and any ‘information’ in them is likely to be misleading, having little connection to reality: On this view, dreams are best ignored. These two views are each widely held in the lay population, and amongst psychologists (see authors included in Pace-Schott et al., 2003).
Dreams can be defined as the images and thoughts that are experienced during sleep. Approximately 14 per cent of people report dreaming every night, and 6 per cent report never dreaming; dream recall frequency decreases with age (Strauch & Meier, 1996). Positive interest in dreams is associated with higher dream recall, but despite many studies over the last decades very few personality correlates of frequency of dream recall have been found (Blagrove, 2007). It is thus unclear why some people report more dreams than others. People who are more creative have been found to recall more dreams, and to report more dreams that are bizarre, but it is not clear if their dream content and actual occurrence of dreams are different from those of less creative people; they may have a greater interest in dreams than less creative people, and thus may just be more willing to report bizarreness or to retrospectively report a high frequency of dreaming (Blagrove, 2007). Individual differences in dream content salience and intensity have been proposed to account for the fact that some people report dreams more frequently than others, but this hypothesis is problematic because the salience of dream content is difficult to assess independently of dream recall.
Dreaming and the brain
Sleep in humans is divided into light sleep (stages 1 and 2), deep sleep (stages 3 and 4), and rapid eye movement sleep (REM sleep). REM sleep occurs approximately every 90 minutes, each REM period becoming longer across the night; the last REM period can be as long as 40 minutes, the first as short as 5 minutes. These eye movements can be visible to a person watching. Brainwaves during REM sleep have some similarity to waking brainwaves, and some parts of the brain are very active, while muscle tone is very low; for this reason, REM sleep has been termed paradoxical sleep.
Aserinsky and Kleitman (1953) found that dream recall was more likely if people were woken from REM sleep than from sleep stages 1 to 4 (non-REM sleep). However, in 1962 Foulkes showed that dreams can occur in non-REM sleep but that they were usually shorter and less dramatic and vivid than REM dreams. It is currently a matter of dispute whether differences between REM and non-REM dreams are just due to non-REM dreams being shorter on average than REM dreams, or due to non-REM sleep not allowing the recall of dreams as well as does REM sleep, or whether there are fundamental differences between the two types of sleep in how they produce dreams (Nielsen, 2000). For example, Hobson et al. (2000) propose a cholinergic neurochemical difference between REM sleep and both wake and non-REM sleep, to explain the amnesic and irrational nature of REM dreams. In various studies since the 1970s Hobson has proposed a physiological basis for other dream characteristics, such as the activation in REM sleep of areas of the parietal lobe being related to visual imagery, activation of the amygdala and paralimbic cortex in REM sleep causing the emotionality of dreams, and random stimulation of the cortex by the brainstem during REM sleep causing scene changes and bizarreness. Hobson’s early activation-synthesis theory of dreaming holds that memories are randomly stimulated by the periodic brainstem activation during REM sleep, with the disparate images being synthesised by the cortex to make a more sensible plot.
However, Solms (1997) points out that whereas non-REM dreams are on average shorter than REM dreams, 20 per cent of non-REM dreams are indistinguishable from REM dreams, and, indeed, many of the characteristics of REM dreams occur also in non-REM dreams. Hobson’s recent work (e.g. Hobson et al., 2000) tries to account for this. Solms shows from his own work and from over a century of the neuropsychological literature that lesions to the ventral-mesial quadrant of the frontal lobe, which is involved with emotional motivation and wishes, or to the parietotemporo-occipital junction (a sensory area), results in the loss of dream recall, but with REM sleep preserved. These brain areas thus seem to be important for creating dreams, with dream production being independent of REM sleep. A debate between Solms and Hobson on whether dreaming is especially associated with REM sleep can be purchased
Specific emotions or moods occur in 75 per cent of dreams, with positive and negative emotions occurring equally often across individuals, joy being the most common single emotion, and anger and fear as next most common (Strauch & Meier, 1996). When an unselected sample of all dreams over a period is collected from an individual, rather than relying on the recollection of particular interesting dreams, or their impression of their dreams in general, bizarreness is found to occur infrequently, in only about 10 per cent of dreams.
Freud’s term ‘day-residue’ refers to the frequent incorporation of events from the previous day into dreams. This is even found in the sleep laboratory, where a third of dreams are found to refer to the laboratory environment and events. Decades of studies on the relationships between dream content and waking life events, cognition and emotions are documented in Kramer (2007); for example, the dreams of people undergoing surgery and psychotherapy, and dreams after divorce. Propper et al. (2007) recorded dreams in the weeks after 9/11, finding dream references to the attacks were a function of the amount of TV coverage of the attacks watched on the day.
Such work looks at the reaction of dream content to waking-life events. Domhoff takes more of a trait approach, quantifying the surface content of large samples of dreams from one person over time, or from a group of people. Many variables can be derived, for example the percentage of characters who are aggressive, the percentage of dreams with negative emotions, the percentage of dreams that contain a misfortune, or the percentage that contain a success. Domhoff shows that these variables can be stable across time. This method of the scientific study of surface dream content, and related statistics, is described on the website www.dreamresearch.net, and in Domhoff (2003). Importantly, the possibility of a motivated selection by participants of what dreams to recount is reduced by having participants recall their Most Recent Dream: this method enables appropriate non-selective sampling of dreams while allowing participants to be required to recall only one dream.
However, although dream content variables can be statistically related to waking-life experiences and other waking-life variables, this seems to account for only a minority of the entire dream content, leading to the claim by Hobson (1999) that much of dream content is delirium-like, rather than motivated or meaningful.
The delirium view of dreams is a more extreme version of the common view that there are deficiencies of cognition in dreams, such as of memory, rationality and volition. We rarely notice bizarre instances, or deliberate about what to do next in the dream, or even realise that it is a dream. One possible physiological parallel noted by Hobson is that deficiencies in volition during dreams might occur due to the lack of activity of the dorsolateral prefrontal cortex during sleep. However, there is evidence from Kahan and LaBerge (1996) that volition, choice and deliberation can be shown to occur in dreams if the dreamer is asked about it soon after waking, rather than with the more usual method of having an independent judge rate written dream reports for the presence of characteristics such as choice, deliberation, self-talk and thwarted intention.
That just 4 per cent of REM dreams are lucid – dreams in which one knows one is dreaming (Gackenback & LaBerge, 1988; LaBerge, 2007) – reinforces the deficiency view of dreaming. However, levels of self-awareness and lucidity in dreams can be increased by cognitive training methods, and Blagrove and Hartnell (2000) found lucid dreamers have a greater internal locus of control and need for cognition than do non-lucid dreamers, which is evidence for the continuity hypothesis of dream variables being associated with waking-life variables.
Dreaming and insight
A frequent question is whether dream content provides information about waking cognition and emotions that is not within conscious awareness, and whether studying dreams can thus provide insight – even therapeutic insight. However, a complication here is that insight could be obtained from dreams in the same way that insight can be obtained from the reading of Tarot cards, or horoscopes, with the process and effort of interpretation providing the reorganisation and linking of information. To address this, attempts have been made to compare the insight obtained from dream interpretation with that obtained by interpreting someone else’s dream, or interpreting an event of one’s own life. However, the results are unclear, and are no doubt confounded by participants’ beliefs about whether dreams are a superior source of insightful knowledge.
One possibility here is to assess claims that dreams can provide the inspiration for inventions and discoveries, such as is claimed for the discovery of the structure of the benzene molecule by Kekulé; the plot for Dr Jekyll and Mr Hyde by Robert Louis Stevenson; the machine sewing needle (with a hole near its tip) by Elias Howe; and Stockhausen’s Helicopter Quartet. However, some of these anecdotal claims are disputed (see www.bps.org.uk/dream and Blagrove, 1992), and hence there always remains the possibility that the surprise in looking at dreams can be due to the imagery they use to depict what we know already, rather than due to any surprise at new knowledge.
Freud (1900/1953) claimed that dreams provide the ‘royal road to the unconscious’, because during sleep we censor thoughts less than when awake. His method of dream interpretation required the person undergoing analysis to free-associate to each component of the dream (see box opposite), the theory being that this would lead from the manifest dream content back to the latent dream thoughts (the wishes and other factors, often unconscious, that were the source of the dream). Clearly this method can result in confabulation of links between the dream content and waking-life cognition and memories, and such criticisms were made at the time of the publication of Freud’s The Interpretation of Dreams, and in the major critique by Grünbaum (1984).
A similar free-association method is, however, now used by Cavallero and many other Italian researchers to distinguish whether items in dreams from the different stages of sleep have as their source a waking episodic memory, a semantic memory, or semantic knowledge about the self.
One possible mechanism for insight from dreams follows from the results of Wegner et al. (2004), where thoughts experimentally suppressed during the day returned during dreams. It may be that with reduced focused thought and less executive control of thinking during sleep, more remote associations can be made. Hartmann (1995) claims that dreams can thus provide a form of therapy without a therapist, although a cognitive method when awake is then needed for altered cognition and action to result from that insight (Hill, 2003).
Does dreaming have a function?
Even if dreams can provide insight, it does not follow that evolution has selected for dreaming. Such insight could occur as an unselected by-product of a functionless residual and deficient mentation during sleep. Whether dreams have an information-processing or emotion-processing function is currently a matter of vigorous dispute. Flanagan (2000) claims that dreams are an epiphenomenon, as did Hobson’s activation-synthesis theory, but against this view are two types of information-processing theories of dreams. The first sees dreams as a kind of ‘virtual reality’ environment in which we can practise waking activities such as of threat simulation and avoidance (Revonsuo, 2000). The second type sees dreams as loose thinking, the making of connections between memories that would not be linked by focused waking thinking alone, or the making of links between emotions and memories (Stickgold et al., 2001).
There is much evidence that various stages of sleep are involved in memory consolidation (Maquet et al., 2003; Van der Werf, 2009), and so it may be that dreams are involved with this function of forming novel connections between memories. Any such function would have to hold even for dreams that are not thought about or recalled on waking (which is the vast majority of our dreams). It may even be that this forgetting and non-recall of most dreams is necessary for any memory consolidation function. However, there are problems for the linking of dream and REM sleep functions. Firstly, various memory-consolidation functions have been found for non-REM sleep, where dream recall is less common; and secondly, REM sleep has mainly been associated with procedural memory learning, rather than declarative learning, which is not easily reconcilable with REM dream characteristics. In favour of a memory function, though, are the findings by Nielsen et al. (2004) and Nielsen and Stenstrom (2005) of the dream lag effect: memories from one day before tend to be preferentially incorporated into dreams, but so do memories from 5–7 days ago. Nielsen proposes that this reappearance of memories in dreams is part of a process of slow consolidation of memories.
In conclusion, the nature of dreaming and of dream content has been extensively studied experimentally since the discovery of an association between REM sleep and dreaming in 1953. This research area has implications for the scientific study of consciousness, and for theories of the possible functions of sleep and of the individual stages of sleep. There is now a vast literature on this field, as reviewed in the three-volume work The New Science of Dreaming (Barrett & McNamara, 2007) and the APA journal Dreaming. However, the two main questions, whether dreaming has a function versus being epiphenomenal, and whether dream content can give insight into cognitions of which we are not consciously aware, are both unsolved. These are real questions, and they are experimentally tractable. That’s why this field remains as exciting as when its pioneers in the 1950s stayed awake in sleep laboratories in order to capture their participants’ dreams.
Some studies have shown that people with frequent nightmares have lower well-being or anxiety in their waking lives (Hartmann, 1998), and there are certainly traumatic nightmares that may occur early in the night and repeat waking-life trauma (Barrett, 1996). However, other research finds only weak relationships between nightmare frequency and current waking-life psychopathology, leading to the possibility that it is the degree of distress caused, rather than the number of nightmares, that is related to psychopathology (Blagrove et al., 2004).
This also raises the possibility that nightmares can be common even in unstressed non-anxious people, and even that we seek negative experiences in our dreams. Indeed, some psychologists (e.g. Revonsuo, 2000) see dreams as an environment for the virtual-reality practising of waking activities, such as of threat simulation and avoidance.
A dream interpreted
At the risk of self-indulgence, I relate the following dream, from the morning of 12 July 2008, the last day of the 25th Conference of the International Association for the Study of Dreams, in Montreal. It illustrates the use of free-association, and the sudden realisations that can occur during this process.
The dream: I am standing. Julia, my partner, walks up to me and gives me a collection of two or three CDs. Julia walks away, this is all done like a present/gift, very nice. I look down and on the top CD is a Rembrandt-like (self-) portrait, with a big floppy hat, looking proud and calm. The portrait may be part of the CD, or just stuck on top of it. The CD had written on it, either on the side or above the portrait, either Rembrandt or Rembrandts. In the dream I wonder if this is the same Rembrandts as did the theme tune to Friends. I think I then wonder what the music is like, as I don’t know it.
I met with others from the conference in a dream interpretation group shortly after waking. At the start of each day of presentations at the conference there are dream interpretation groups led by more therapy-inclined colleagues. I was reticent about the dream because it was so short and obviously meaningless, but reported it anyway.
The group first asked for clarifications of the dream.
I I stated that the portrait was like a Dutch master, with black and brown and greys, very intricate.
I A man asks whether the hat is like a professorial hat. I realise that I will be wearing such a hat next week, at our university graduation ceremony, and that this image may refer to my recent promotion to Professor and Head of Department.
I A woman asks about Julia giving something creative to me. I realise this refers to her having been an artist, a profession which she gave up, and also that she has given up so much to allow me to get promotion. She is also giving me the two or three CDs, just as she has given us our two children, and, like in the dream, there is no resentment, all is calm and happy.
I I now associate to one of yesterday’s conference sessions. In this session
I harshly questioned a very orthodox Freudian psychoanalyst about her belief in the use of free-association, asking her what she thinks of Grünbaum’s sceptical work on this. She did not understand my question. Her talk had been about the presence of condensations in dreams, where one part of a dream can be associated to many parts of waking life, and also about how inconsequential minor characters from one’s past can suddenly appear in dreams. I realise this is like the appearance of Rembrandt in the dream, a minor character whom I rarely think of.I find this amazing, it is like my question to her about the validity of free-association was answered by a dream, which had so many condensations and a figure from the past.
I I then associate Rembrandt with REM!
I I associate Rembrandts and the theme tune of Friends to the friendly interaction between Julia and me in the dream, as in waking life.
I I associate the picture to a 13-year-old, Ben, to whom I am a very close uncle, and who now lives in the Netherlands: a ‘Dutch master’!
I I associate Rembrandt to Remember: last night I ate with people from the Montreal Sleep Lab who talked of the Quebec car licence plate phrase Je me souviens, ‘I remember’. I wonder aloud whether the dream is telling me to remember all that Julia is doing and has done for me. I had indeed thought about that the day before, and was thinking of how she had given me the time to be at the conference.
I Final association: The previous evening an academic told me that he had become a head of department and that I should see my new post like having my own ‘palette’, creatively producing a painting of the department.
One possibility to account for the above associations is that, as Adolf Grünbaum claims, it is all confabulation. Another possibility is that overlooked thoughts of gratitude towards my partner found metaphorical expression during the dream. How those two possibilities can be experimentally tested is a major question in dream research.
Aserinsky, E. & Kleitman, N. (1953). Regularly occurring periods of ocular motility and concomitant phenomena during sleep. Science, 118, 361–375.
Barrett, D. (Ed.) (1996). Trauma and dreams. Cambridge, MA: Harvard University Press.
Barrett, D. & McNamara, P. (Eds.) (2007). The new science of dreaming: Vols 1–3. Westport, CT: Praeger.
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, 205–219.
Blagrove, M. (2007). Dreaming and personality. In D. Barrett & P. McNamara (Eds.) The new science of dreaming: Vol 2. Content, recall, and personality correlates. Westport, CT: Praeger.
Blagrove, M., Farmer, L. & Williams, E. (2004). The relationship of nightmare frequency and nightmare distress to well-being. Journal of Sleep Research, 13, 129–136.
Blagrove, M. & Hartnell, S. (2000). Lucid dreaming: Associations with internal locus of control, need for cognition and creativity. Personality and Individual Differences, 28, 41–47.
Domhoff, G.W. (2003). The scientific study of dreams: Neural networks, cognitive development, and content analysis. Washington, DC: American Psychological Association.
Flanagan, O. (2000). Dreaming souls. Oxford: Oxford University Press.
Foulkes, D. (1962). Dream reports from different stages of sleep. Journal of Abnormal and Social Psychology, 65, 14–25.
Freud, S. (1953). The interpretation of dreams. In J. Strachey (Ed. and Trans.) Standard edition of the complete psychological works of Sigmund Freud (Vols 4–5). London: Hogarth Press. (Original work published 1900)
Gackenback, J. & LaBerge, S. (Eds.) (1988). Conscious mind, sleeping brain. New York: Plenum Press.
Grünbaum, A. (1984). The foundations of psychoanalysis: A philosophical critique. Berkeley, CA: University of California Press.
Hartmann E. (1995). Making connections in a safe place. Dreaming, 5, 213–228.
Hartmann, E. (1998). Dreams and nightmares: The new theory on the origin and meaning of dreams. New York: Plenum Press.
Hill, C.E. (Ed.) (2003). Dream work in therapy. Washington, DC: American Psychological Association.
Hill, C.E., Diemer, R., Hess, S. et al. (1993). Are the effects of dream interpretation on session quality, insight, and emotions due to the dream itself, to projection, or to the interpretation process? Dreaming, 3, 269–280.
Hobson, J.A. (1999). Dreaming as delirium. Cambridge, MA: MIT Press.
Hobson, J.A. Pace-Schott, E. & Stickgold., R. (2000). Dreaming and the brain: Towards a cognitive neuroscience of conscious states. Behavioral and Brain Sciences, 23, 793–842.
Kahan, T. & LaBerge, S. (1996). Cognition and metacognition in dreaming and waking: Comparisons of first and third-person ratings. Dreaming, 6, 235–249.
Kramer, M. (2007). The dream experience. New York: Routledge.
LaBerge, S. (2007). Lucid dreaming. In D. Barrett & P. McNamara (Eds.) The new science of dreaming: Vol 2 (pp.307–328). Westport, CT: Praeger.
Maquet, P., Smith, C. & Stickgold, R. (2003). Sleep and brain plasticity. Oxford: Oxford University Press.
Nielsen, T. (2000). Cognition in REM and NREM sleep. Behavioral and Brain Sciences, 23, 851–866.
Nielsen, T.A., Kuiken, D., Alain, G., et al. (2004). Immediate and delayed incorporations of events into dreams. Journal of Sleep Research, 13, 327–336.
Nielsen, T.A. & Stenstrom, P. (2005). What are the memory sources of dreaming? Nature, 437, 1286–1289.
Pace-Schott, E.F., Solms, M., Blagrove, M. & Harnad, S. (Eds.) (2003). Sleep and dreaming: Scientific advances and reconsiderations. Cambridge: Cambridge University Press. (Reprint of a special issue on dreaming in Behavioral and Brain Sciences, 2000.)
Propper, R.E, Stickgold R., Keeley, R. & Christman, S.D. (2007). Is television traumatic? Psychological Science, 18, 334–340.
Revonsuo, A. (2000). The reinterpretation of dreams. Behavioral and Brain Sciences, 23, 877–901.
Solms, M. (1997). The neuropsychology of dreams. Mahwah, NJ: Lawrence Erlbaum.
Stickgold, R., Hobson, J., Fosse, R. & Fosse, M. (2001). Sleep, learning, and dreams. Science, 294, 1052–1057.
Strauch, I. & Meier, B. (1996). In search of dreams. Albany, NY: State University of New York Press.
Van der Werf, Y.D., Altena, E., Schoonheim, M.M. et al. (2009). Sleep benefits subsequent hippocampal functioning. Nature Neuroscience, 12, 122–123.
Wegner, D.M., Wenzlaff, R.M. & Kozak, M. (2004). Dream rebound: The return of suppressed thoughts in dreams. Psychological Science, 15, 232–236.
BPS Members can discuss this article
Already a member? Or Create an account
Not a member? Find out about becoming a member or subscriber