The Interpretation of Dreams

Home | Annotated bibliography | Other essays

The bizarre, irrational nature of dreams, where reality gets warped and laws of nature are turned upside down, and why we remember them at all, are some of the most puzzling mysteries of the mind. Dreaming is a process that absorbs a lot of energy; therefore, it must serve a purpose, possible an important one.

Freud's 1900 theory of dreams stood on the following principles: 1. Dreams are composed of sensory images; and 2. Free associations are evoked in the dreamer's mind by these images. He concluded that dreams rely on memories and that they are assembled by the brain to deliver a meaning. Meaning of dreams are hidden and reflect memories of emotionally meaningful experiences.

Freud's work had an unfortunate consequence on neuroscience: scientists became more interested in the "content" of dreams than in the "form" of dreaming. Scientists were looking for the "meaning" of dreams, rather than for the "source" of dreams. Scientists studying dreams behaved like doctors analyzing symptoms of a desease, rather than like physicists looking for the causes of a natural phenomenon. This historical accident basically caused dreams to remain outside the sphere of science for seven decades.

Much more important was a finding that remained neglected for almost a century: at the end of the 19th century the British neurologist John Hughlings Jackson realized that a loss of a brain function almost always results in the gain in another brain function. Typically what is gained is heightened sensations and emotions. Jackson, virtually a contemporary of Darwin, explained this phenomenon with the view that the brain's functions have different evolutionary ages: newer ones took over older ones, but the older ones are still there, we just don't normally need to use them as the newer ones are more powerful. When we lose one of the newer features, then the older features of the brain regain their importance. Jackson had the powerful intuition that a single process was responsible for a "balance" of brain states.

An important discovery (probably the one that opened the doors of the neurobiology of dreams) occurred in 1953: during sleep, the brain enters a state of "rapid eye movement". It turns out that this is the state in which dreaming occurs. REM sleep recurs regularly. A brain enters REM sleep 4 or 5 times per night, at approximately 90 minute intervals, and each period lasts about 20 minutes.

In 1962 the French physiologist Michel Jouvet observed that REM sleep is generated in the pontine brain stem (or "pons"). In other words, Jouvet localized the trigger zone for REM sleep and dreaming in the brain stem.

REM sleep exhibits four main properties:

    • A low level of brain activity
    • Inhibition of muscle tone
    • Waves of excitation from the pons

    • Rapid eye movement

The waves of excitation are probably the cause of everything else. The pons sends signals and excites eye muscles (causing rapid eye movement), the midbrain (causing a low level of brain activity), and the thalamus. The thalamus contains structures for visual, auditory, tactile and so forth cognition. The thalamus then excites the cortex. The cortex therefore receives a valid sensory signal from the thalamus and interprets it as if it were coming from the eye, ears, etc.

During REM sleep several areas of the brain are working frantically, and some of them are doing exactly the same job they do when the brain is awake. The only major difference is that the stimuli they process are now coming from an internal source rather than from the environment: during dreams the sensory input comes from the sensory cortex.

Dreams are made of this

There are three main categories of explanation for dreams. The simplest explanation is that dreams are just an evolutionary leftover. By accident we have five fingers rather than four. By accident we dream while we sleep. Another explanation is that they are fossils of a previous form of mind, accidental remnants of previous brain processes. Yet another explanation is that they are a window on some kind of processing that goes on in the brain while we sleep.

Imagine that somebody is filing a lot of newspaper clippings in folders and are standing in front of him: you will see a rapid sequence of titles flashing in front of your eyes. While you understand each of them, the flow of titles is cryptic: it may form, by mere chance, stories, but stories that you cannot understand. In reality the sequence of titles is not random, because the person who is filing the titles is following a logic (for example, they are filed in chronological order, or in order of importance, or by subject matter). It is just that you are only a spectator. This could be exactly what is happening to our consciousness while we are sleeping. The brain is rapidly processing a huge amount of information in whatever order and our consciousness sees flashes of the bits that are being processed. These bits seem to compose stories of their own, and no wonder that the stories look weird if not undecipherable.

This third hypothesis is based on neurophysiological findings. The brain, far from being asleep, is very active during sleep. Most nerve cells in the brain fire all the time, whether we are awake or asleep.

There is growing consensus among neurobiologists that remembering and forgetting occurs during dreams.

Traditionally, neurophysiologists have studied brain activity during sleep, and neglected its awake states. But it turns out that they are surprisingly similar.

Dreams are made for this

Jouvet was also a pioneer of the theory that dreams have a function: to derive crucial action patterns from the genetic program of the individual. REM sleep provides a means to combine genetic instructions with experience. Sleep and dreaming are a survival strategy.

According to his findings, a dream is the vehicle employed by an organism to cancel or archive the day's experiences on the basis of a genetic program. This explanation would also reconcile the dualism between hereditary and acquired features: how much of what we know is innate and how much is acquired by experience? In Jouvetís scenario, an hereditary component is activated daily to decide how new data must be acquired.

The American neurobiologist Jonathan Winson expressed this concept in a more general way: dreams represent "practice sessions" in which animals (not only humans) refine their survival skills.

REM sleep helped the brain "remembering" important facts without having to add cortical tissues. During REM sleep the brain (specifically, the hippocampus) processes information that accumulated during the day. In particular, during REM sleep the brain relates recent memories to old memories, and derives "tips" for future behavior. Dreams are a window on this "off-line processing" of information.

First of all, amingergic neurotransmitters populate the brain. This is the precondition to the "theta rhythm" of electrical activity in the hippocampus. Such theta rhythm has the effect of inhibiting the passage of signals to the limbic system. Winson emphasizes that theta rhythm occurs only in two circumstances: 1. States that are important for survival (that are normally specific to each species) and 2. REM sleep (dreaming).

Winson's hypothesis is that, during sleep, the hippocampus processes the day's events and stores important information. Winson postulates a strong connection between dreaming (or whatever causes dreaming) and long term memory.

Dreaming is an accidental feature that let us "see" some of the processing, although only some: a dream is not a story but a more or less blind processing of the day's experience.

There is, therefore, a biologically relevant reason to dream: dreaming is a sort of off-line processing essential to learning to survive in our environment.

Freud was right that dreams are the bridge between the conscious and the unconscious, although that bridge is of a different nature. The Freudian "subconscious" becomes the phylogenetically ancient mechanism involving REM sleep, in which memories and strategies are formed in the prefrontal cortex.

In 1983 Francis Crick also proposed that the function of dreams is to "clear the circuits" of the brain, otherwise there would not be enough space to register each day's events.

We can summarize these ideas as follows. The brain, in the face of huge daily sensory stimulation, must:

    • understand what matters
    • understand what does not matter
    • remember what will still matter
    • forget what will never matter again

Dreams help out.

The ultimate purpose of dreams is to help us learn. We dream hypothetical situations so that we will be prepared to face real situations of the same kind. When a waking situation occurs, it has probably already been played at least once in our dreams, and we know what to expect. By dreaming, we train our brain: dreams are mental gymnastics. It's like saying that, in order to see something, we must first create the vision of that something in our mind.

What is still missing is the physical link between dreams and genome. Neurotransmitters (such as animenes and cholines) act on the surface (the membrane) of the cell, whereas genes lie in the center (the nucleus) of the cell. But the messenger molecules transfer information from the membrane to the nucleus and viceversa. Hobson has hypothesized that neurotransmitters may interact with messenger molecules and therefore affect the work of genes.

Whether driven by the genetic program or not, what the brain does during sleep is consolidating memories that have been acquired during the day. Dreaming, far from being an eccentric manifestation of irrationality, is at the core of human cognition.

Dreams are made of consciousness

Whether sleeping or awake, the brain does pretty much the same thing. The dreaming brain employs the same systems and processes of the awake brain, except that those processes are not activated by stimuli from the outside world; that the outcome of those processes does not result in (significant) body movements; and that self-awareness and memory are dormant. The American psychiatrist Allan Hobson summarized it as: the input, the output, the processor and the working space of the awake brain are replaced by something else.

What makes a difference is the neurotransmitters that travel through the brain. What differs between wake and sleep is very little, but enough to alter dramatically the outcome: during sleep the brain is bombarded by erratic pulses from the brain stem and flooded with nervous system chemicals of a different sort.

Neurotransmitters make brain circuits more or less sensitive. Aminergic neurotransmitters originate in the brain stem and terminate in the amygdala cholinergic neurotransmitters originate in the forebrain and terminate in the cortex. During waking states, the brain is controlled by the aminergic neurotransmitters, made of molecules called "amines". During sleep, the brain is controlled by the cholinergic neurotransmitters, made of a molecule called "acetylcholine". Cholinergic chemicals free the system used for cognition and behavior. They paralyze the body by sending pulses to the spinal chord, even if motor neurons are always in motion.

These two chemical systems are in dynamic equilibrium: if one retracts, the other advances. This means that our consciousness can fluctuate between two extremes, in which either of the chemical systems totally prevails (neither is ever completely absent). This also means that the brain states of wake and sleep are only two extremes, between which there exists a continuum of aminergic-cholinergic interactions, and therefore a continuum of brain states. This system can be said to control the brain. It resides in the brain stem and from there it can easily control both the lower brain (senses and movement) and the upper brain (feelings and thought).

When it doesn't work properly, when the balance of chemicals is altered, mental diseases like delirium occur. It is not surprising that diseases such as delirium are so similar to dreams: they are driven by exactly the same phenomenon.

Hobson claims that the brain is in awake, dream or (non REM) sleep mode depending on whether amines are prevailing, cholines are prevailing or amines and cholines are "deadlocked".

Three factors account for the brain behavior at any time: activation energy (amount of electrical activity), information source (internal or external) and chemical system (amines or cholines).

When activation energy is high, the information source is external and the mode is aminergic: the brain is awake. As activation energy decreases, the external information source fades away and amines and cholines balance each other: the brain falls asleep. When activation energy is high, the information source is internal and the mode is cholinergic: the brain is dreaming. During an hallucination: activation energy is high, the information source is internal and the mode is aminergic. In a coma: activation energy is low, the information source is internal and the mode is cholinergic.

The extremes are rare and usually traumatic. Normally, both external and internal sources contribute to the cognitive life, and both amines and cholines contribute to the brain state.

The interplay of external and internal sources means that our perceptions are always mediated by our memory. Hobson thinks that our brains do not merely react (to stimuli), they also anticipate. The internal source tells us what to expect next, and thus helps us cope with the external source. Emotions are, in a sense, a measure of how well the internal source matches the external source: anxiety is caused by a major mismatch, whereas contentness is a sign of matching sources.

When we dream, the spinal cord is paralyzed and the senses are disconnected. This is because of the cholinergic neurotransmitters that come from the brain stem.

Hobson believes that sleep has the function to reinforce and reorganize memory: ultimately, to advance them from short-term memory to long-term memory. Amines are necessary for recording an experience, cholines consolidate memory. Hobson deduces that during REM sleep memory is consolidated.

The aminergetic system is responsible for attention, focus, awareness. The cholinergetic system is responsible for the opposite process: focus on nothing, but scan everything.

As for the content of dreams, Hobson thinks that they reflect a biological need to keep track of place, person (friend, foe or mate) and time. He draws the conclusion from considerations about what is typical (and bizarre) of dreams: disruptions in orientation.

The bottom line is that dreams are meaningful: the mind makes a synthetic effort to provide meaning to the signals that are generated internally (during a dream, memory is even "hypermnesic", i.e. is intensified). Wishes are not the cause of the dreaming process, although, once dreaming has been started by the brain stem, wishes may be incorporated in the dream. Therefore, Hobson thinks that dreams need not be interpreted: their meaning is transparent. Or, equivalently, dreams must be interpreted in the realm of neurophysiology, not psychology.

The interplay between the aminergic and the cholinergic systems may be responsible for all conscious phenomena (for Hobson, dreams are as conscious as thinking) and ultimately for consciousness itself. After all, conscious states fluctuate continuously between waking and dreaming.

Dreams, far from being subjective, are "impersonal necessities forced on brain by nature".

An evolution necessity

The American psychiatrist Fred Snyder was the first one (in the 1960s) to advance the notion that, from an evolutionary perspective, REM sleep came first and dreams came later. First bodies developed the brain state of REM sleep, which was retained because it had a useful function for survival (for example, because it kept the brain alert and ready to react to emergencies even during sleep), and then dreams were engrafted upon REM sleep. REM sleep was available and was used to host dreams. Dreaming evolved after a physical feature made them possible, just like language evolved after an anatomical apparatus that was born for whatever other reason. Dreaming, just like language, is an "epiphenomenon".

Anthony Stevens has provided a practical explanation for why some animals started dreaming: dreaming emerged when oviparous animals evolved into viviparous animals. By dreaming, the brain could augment its performance with some "off-line" processing. This made possible to limit the size of the brain while leaving brain activity free to grow. Brains, and thus heads, would remain small enough to pass through the maternal pelvis.

In Winson's scenario, dreams helped us survive a long time before our mind was capable of providing any help at all. And dreams, unlike higher consciousness, are likely to be common to many species.

The mind could well be an evolution of dreaming, which happened in humans and not in other species. First the brain started dreaming, then dreams took over the brain and became the mind, which could be viewed as a continuous dream of the universe.

This hypothetical history of the mind does not differ too much from the one in which the mind was created by memes. The relationship between memes and dreams is intuitive, and psychologist Joseph Campbell indirectly summarized it with his celebrated aphorism that "a myth is a public dream, a dream is a private myth".

Of how real dreams are and how dreamy reality is

The experience of a dream may feel so utterly bizarre for today's mind, but we have to go back millions of years to realize that it is probably far less bizarre than it appears to us today. It is likely that millions of years ago our waking life was not too different from our dreaming life. Consciousness in dreams is a series of flashes which are fragmented and very emotional. It is likely that awake consciousness had exactly the same character: mostly nothing would happen to our consciousness (no thinking, no emotions, just mechanic, instinctive behavior) but situations would present suddenly that would arouse strong feelings and require immediate action. Our awake life "was" a series of emotionally charged flashes, just like dreams. The difference between being awake and dreaming was only the body movement. As we rehearsed the day's events during dreams, we would feel that the sensations are perfectly normal.

Today our consciousness has acquired a different profile: it has evolved to a more or less smooth flow of thoughts, in which strong emotions don't normally figure prominently. We think when we are commuting on a bus or while we are shopping in the mall, and the most violent emotion is being upset about the price of a shirt or suddenly realizing we just missed our stop. They are peanuts compared with the emotion of being attacked by a tiger or of being drawn by strong currents towards the waterfall. Our awake consciousness has changed and dreams have remained the same. The brain is still processing off-line, during sleep, our day's events with the same cerebral circuits that we had millions of years ago, and therefore it is still generating the same flow of emotionally-charged flashes of reality. When the brain is awake, reality does not impinge on those circuits in the same way it did in the hostile, primitive environment of million of years ago. The world we live in is, by and large, friendly (free of mortal foes and natural catastrophes). But when danger does appear (a mortal foe or a natural catastrophe), then our awake life becomes just like a dream: "it was a nightmare", "it didn't feel real", etc. In those rare and unfortunate circumstances (that hopefully most of us will never experience) our waking life feels just like a dream: flashes of reality, violent emotions, apparent incoherence of events, etc.

Because of the society that we have built and the way we have tamed and harnessed nature's unpredictability through civilization, our brain does not receive the sudden and violent stimuli it used to. This is what makes most of the difference between being awake and dreaming. It is not a different functioning of the brain, it is a different functioning of the world around us.


What have joking and dreaming in common? Apparently nothing, but they both belong to the category of acts that do not seem to have a useful funtion. Like dreaming, joking seems to be a pointless waste of energies. Like dreaming, joking is some kind of playing with our experience. Like dreaming, joking is process of rearranging our experience in a somewhat irrational way. Like dreams, jokes do not necessarily require linguistic skills, but normally occur in a linguistic context. More than dreams, actually, jokes tend to rely on language. More than dreams, jokes seem to have developed in humans to a level far more sophisticated than in any other species. We see animals play and laugh, but the gap between a comedian and two lion cubs wrestling in the grass is enormous.

First, we may want to ponder whether human dreams too are so much more complex than other speciesí dreams. Second, we may want to ascribe this complexity to the acquisition of language.

Third, we may want to use what we know about dreams to explain why we make jokes at all. While there is no biological evidence to support the idea that jokes have a specific function for our learning and survival, one wonders why we enjoy so much making them. Woody Allen once said that comedy is tragedy plus time: when something tragic occurs, it is inappropriate to make fun of it, but months or years later it may be perfectly appropriate. If I trip on something and break my leg, I am in no mood to hear a joke about it, but it is more than likely that years later somebody will mock me on this subject. Jokes refer to past experience, and usually refer to tragic experience. If not tragic, then significant in some way. The point is that, indirectly, jokes help us learn and remember.

The Evolution of Memories

Memory is not a storage, because it cannot recall events exactly the way they were. Memories change all the time, therefore memory is not a static system, it is a dynamic system.

Memory is pivotal for the entire thought system of the individual. Therefore, memory is about thought, it is not limited to remembering. Memory stores and retrieves thoughts.

Memory can be viewed as an evolving population of thoughts. Thoughts that survive and reproduce are variations of original thoughts, and somehow "contain" those original thoughts, but adapted to the new circumstances. Memories are descendents of thoughts that occurred in the past. Thoughts are continuously generated from previous one, just like the immune system generates antibodies all the time and just like species are created from previous ones.

Memory, far from being a static storage, is changing continuously. It's not a location, it is the collective process of thinking. Every thought contains memories.

Further Reading

Hobson J. Allan: THE DREAMING BRAIN (Basic, 1989)
Hobson Allan: THE CHEMISTRY OF CONSCIOUS STATES (Little & Brown, 1994)
Jouvet Michel: LE SOMMEIL ET LE REVE (Jacob, 1992)
Snyder Frederick: EXPERIMENTAL STUDIES OF DREAMING (Random House, 1967)
Winson Jonathan: BRAIN AND PSYCHE (Anchor Press, 1985)


Back to cognitive science