Back to the blog

Blog 14: Dream associations, the unconscious and modern dream science

30th September 2020

What is the evidence from modern science on unconscious dream associations? Here I refer to evidence on two questions. First, do dream associations come from the unconscious? And, second, do dream associations go to the unconscious.  But we should note that dreaming itself is an active, conscious state. If you are awakened from a dream you can report your conscious experience of what dream was about.  The phenomenon of dream consciousness blew away the notion that all of sleep is passive and unconscious. Also the dreamer (and others) can use dream reports to identify associations.  But, unfortunately, it is more or less impossible to discover what people are dreaming about while they are asleep and dreaming.

Although, science is making progress on “dream-reading machines”. A team of Japanese researchers devised a way to, broadly, tell, what images people were seeing in their dreams. For example,  the researchers could tell someone was dreaming of a building but couldn’t narrow this down to the structure being a house.  Broad categories, like building, animal and person were consistent with participants’ dream reports.

What is she dreaming of?

But, at the moment, for meaningful and detailed dream reports, we are reliant on waking people up from sleep in laboratories or collecting dreams written down at home. A recent study calculated that 96.4% of awakenings from REM sleep produced a dream report. We look at dream reports in the next blog but, first, we review experiments done in rapid eye movement (REM) sleep, when most dreams occur.

First, what is the evidence on whether dream associations come from the unconscious? Recent research supports the Freudian idea of repressed, unconscious memories reappearing during REM sleep because REM sleep makes the recall of unwanted memories easier – where “unwanted” is defined as repressed during wakefulness.  Dreams are composed from memory fragments, which are integrated into a dream narrative. So it seems that at least some dream associations come from repressed memories in the unconscious.

But dream associations could be made unconsciously without being composed from repressed memories. We now know much more about the unconscious than in Freud’s time. As pointed out in an earlier blog, 95% brain activity is unconscious. The influence of repressed, unconscious memories is only likely to drive a small aspect of this brain activity. But, introspection indicates that dreams are unconscious in another sense- they are unconscious constructions. We don’t consciously search our memories for elements which are only remotely associated. Dream scenes appear in our consciousness, already formed and, unless we are having a lucid dream, out of our control.

So, in answer to the question of whether dream associations come from the unconscious, we can conclude that some dreams may feature associations to repressed memories but all dreams seem to be constructed unconsciously, although once formed we consciously experience them.

Second, what is the evidence on whether dream associations go to the unconscious? Modern science also indicates that associations made during REM sleep are retained unconsciously.

Dreams don’t die?


I don’t think dreams die, after we dream them they go to the unconscious.

Researchers at the Weizmann Institute of Science demonstrated unconscious retention for associations made during REM sleep. While their participants were asleep, they played a number of tones,

Hearing tones while asleep

which they followed by odours.

Sniffing deeply 

They showed how, during REM sleep, their participants formed associations between each of the sounds and the odours that followed them.

If an odour is pleasant, like the lavender in the photo above, when we sniff, we sniff deeply.

But if the odour is unpleasant, like the garbage below, when we sniff, we sniff shallowly to protect ourselves from noxious smells.

Sniffing shallowly

In the Weizmann Institute of Science experiment, while their participants were in REM sleep, each of the tones was followed by a pleasant or unpleasant odours.  After these pairings, if only the tones were presented, the participants, again whilst still asleep, sniffed deeply if a pleasant odour was anticipated and shallowly if they expected a noxious smell.

The following day the participants did not consciously remember the tones or the odours or the associations they has made between them during REM sleep. But, the researchers showed that their participants had retained these associations unconsciously because the associations created in sleep impacted on their behaviour during wakefulness. Specifically, on the following day, the participants’ sniffing behaviour showed the same pattern as it had during sleep. When a tone was played which had been associated with a pleasant smell the participants sniffed deeply. Conversely, when a tone was played which had been associated with a unpleasant smell the participants sniffed shallowly.

Importantly, this showed their expectations and, consequent behaviour was influenced by new associations learned during REM sleep- despite their having no memory of these associations. In other words, the behaviour of the participants in wakefulness was driven by unconscious associations formed in REM sleep. This Weizmann Institute of Science experiment cannot prove that these unconscious associations were made during REM dreams but, because of the very high percentage of REM sleep (up to 96.4%) that features dreams, we can assume this to be the case- unless proven otherwise.

This experiment demonstrates a significant distinction, which we often overlook, between forgetting and retention. Even if we recall a dream upon waking we usually forget it later in the day, unless it’s particularly emotionally resonant, in which case the dream can last for years. But, as the above experiment shows, although we forget them, we can retain associations, formed during sleep.  These associations persist and continue to act unconsciously on our behaviour in wake.

So, in answer to the question of whether dream associations go to the unconscious, we can conclude, from the above experiment, that associations formed in REM sleep do go to the unconscious.

When considering the distinction between conscious and unconscious it’s important to recognise that some aspects of the unconscious can become conscious.  Consider that at any point in time we are unconscious of almost all of our long term memories, yet a trigger event or the need to recall can easily bring a relevant memory into consciousness. For example, being at a wedding may trigger memories of previous weddings I have attended or seeing a person I know approaching will, hopefully, bring their name to mind.

On the other hand, some memories or knowledge can only be brought into consciousness with difficulty. For example, memories of how to start a car are clearly present in consciousness while learning to drive but, after a time, become largely unconscious as the process becomes automatic. If you attempt to teach someone to drive, you access the processes of starting the car with some difficulty. You have to think: first turn on the ignition, then check the gear stick is in neutral, next press down the clutch, then engage first gear and so on. For a practised driver these processes are mostly unconscious because they have entered what memory researchers call “procedural memory”, in other words unconscious memories of “how to do things”. Unconscious retention makes our actions and responses faster and more efficient.

Some aspects of the unconscious can never enter consciousness. For example, the processes that run your body- your breathing, heart rate and immune system- can’t be conscious.

An interesting question is whether dreams which have never been consciously recalled could enter consciousness. The phrase “breaking a dream” implies that they can- if a trigger brings into mind a fragment of a dream which previously occurred. Unfortunately, at the present state of dream science, I can’t see how evidence could be brought to bear on this “breaking a dream” question…

Leave a Reply

Your email address will not be published. Required fields are marked *