Freud and his book The Interpretation of Dreams have for almost a century dominated thinking on why we dream and whether the images that flash through our brains as we sleep have any inherent meaning. Things are changing, however, as researchers delve into the biochemistry of mental processes and map the brain as it works.
Last week's American Association for the Advancement of Science annual meeting in Anaheim, California, dedicated a session to current research into the dreaming process. Entitled "Psychoanalytic and Neuro-scientific Perspectives on Dreaming", it teamed researchers from a range of institutions who presented papers on what the brain does when it dreams.
One interpretation of the research results suggests that Freud was probably wrong about aspects of his theories, but not about everything. His views might therefore need some revision but are unlikely to be abandoned completely.
Dr Mark Solms, of St Bartholomew's and Royal London Hospital, presented work which contradicts current thinking about the link between REM (rapid eye movement) sleep and dreaming. Studies over several years had made a direct link between REM sleep and dreaming. Between 70 and 95 per cent of subjects woken during REM reported that they had been dreaming, whereas only 5 to 10 per cent of those woken outside of REM reported dreams.
Studies of what parts of the brain were activated during REM and non-REM contradict this however.
While the evidence now shows there is an important link between the two states, REM sleep can occur without dreaming and dreaming can occur without REM sleep.
Moreover, REM was found to be controlled by a primitive part of the brain, the brainstem, while non-REM was controlled by the higher level forebrain.
If these results were proven it would mean "a major paradigm shift in sleep and dream science", Dr Solms said.
The key is the forebrain involvement, the part of the brain linked to conscious activity and thinking. The forebrain activity would provide support for the longstanding psychoanalytic view that dreams have meaning.
Dr Solms's work was supported by studies carried out by Dr Allen Braun of the US National Institutes of Health's Institute on Deafness and Other Communication Disorders.
He used PET scanning to record brain activity and also found that REM was associated with lower order brain areas and dreaming with higher order areas.
He suggested that activation of the visual areas of the brain might account for the visual component of dreams and activation of the limbic area which is associated with emotional response might relate to the "heightened emotionality of dream images".
The hippocampus, which is linked to short term memory processing, was consistent with dreaming's retrieval and incorporation into the images of the day's leftovers.
These regions, he said, also regulated autonomic functions and might account for the characteristic changes in heart rate, blood pressure and respiration during the two sleep states.
Dreams, he said "may subserve wish fulfilment", consistent with psychoanalytic thinking.
While the data would contradict Freud's assumptions that REM sleep allowed the "screening, censorship and symbolic distortion of unconscious material", it could support the view that dream content was not random and meaningless and may instead "constitute direct - albeit distorted - access to unconscious processes". Freud's theories must be modified rather than discarded, argued Dr Ramon Greenberg of Harvard Medical School. Freud's views on screening and censorship were not consistent with the current research findings or with modern psychoanalytic thinking.
"The manifest content of the dream is given greater significance as the special language of the dream," he said. "Consideration of this must influence how we understand and work with dreams, while retaining Freud's most important contribution - namely dreams are meaningful and the royal road to `consciousness'."
