The Dreams of Rapid Eye Movement (REM) Sleep

The identification of particular EEG/EOG signs with a sleep state in which people are likely to be dreaming, has made it feasible to answer a number of questions that had previously been a matter of speculation but were, and still are, perennially asked by ordinary people.

Answers to most of these questions were provided in the 15 years between 1960 and 1975. There are, of course, still mysteries about dreaming, and for some of them perhaps the technology of the dream laboratory on its own will never provide a solution. As Ian Oswald once remarked to me, we know very little about the function of everyday, waking consciousness, so perhaps it is overoptimistic to believe that we can achieve a complete understanding of sleeping consciousness merely because we have reliable physiological indices of when dreams are likely to occur.

Watching the eye movements of subjects during REM sleep and then obtaining reports from them of intense visual imagery, it seems only natural to assume that one has something to do with the other. That is, the intense visual experiences of dreaming during REM sleep are signaled by eye movements, because the eyes are actually scanning the images generated by the dream. However, it has proved extremely difficult to establish the truth or falsehood of this apparently simple statement.

Some dreams are more visually intense than others, and some REM sleep periods have a great many more eye movements than others. (REM sleep may be scored as such when hardly any rapid eye movements are occurring, as long as the record of low-voltage EEG and reduced EMG continues uninterruptedly.) Berger and Oswald (1962b) have established that REM sleep periods characterized by few and small eye movements are accompanied by what they call 'passive' dreaming, with few reports to indicate that a subject might have been scanning. 'Active' dreams are accompanied by profuse and rapid eye movements. Even within a period of REM sleep there are phases of intense eye movement separated by intervals of ocular quiescence. Molinari and Foulkes (1969) have found that subjects woken up after bursts of eye movement, report more vivid imagery than when woken up after intervals of REM sleep in which the eyes have been still.

The high correlation of bursts of rapid eye movements with intense visual imagery tended to encourage early investigators' impression that the eyes were following the actions in 70-80 per cent of vivid dreams (Roffwarg et a I., 1962). However, when blind matches of EOG records with dream narratives have been made simply on the basis of written evidence and the polygraphic record, the odds on matching reports to records have been no better than chance (Moscovitz and Berger, 1969).