Are Dreams Confined to REM Sleep?

In the 1950s, Aserinsky and Kleitman (1953) and Dement and Kleitman (1957) stated that 74 percent and 79 percent of their subjects reported dreams after REM sleep awakenings, respectively and only 7 percent after non-REM sleep awakenings, indicating a virtually perfect isomerism between the experience of dreaming and a well-defined psycho physiological pattern. In the first flush of enthusiasm following these unprecedented claims, there was a very strong tendency to identify dreaming exclusively with REM sleep, presuming that any lack of identification of dreaming with eye movements could be put down to experimental error. Some researchers, such as Ernest Hartmann, even called it D sleep (the D is representative for dreaming) to distinguish it from non-dreaming, slow-wave sleep. However, it was not long before other researchers were reporting dream recalls after non-REM sleep awakenings, and a lively controversy began on the question of whether night-time dreaming was confined to REM sleep, or whether it might be continuous through all sleep stages.

A thorough review of the literature on this topic by Herman et al. (1978) found that reports of dreaming varied between studies from 74-100 per cent for REM sleep awakenings, and from 0-75 per cent for non-REM sleep awakenings. There were no obvious differences between the ways in which the studies had been conducted, to account for the wide range of rates for non-REM sleep. The studies only agreed on a consistently high rate of dream reports for REM sleep. While sleep stages are well denned in terms of psycho-physiology, the same is not true of what constitutes a dream, and the reviewers suggest that the predisposition of the experimenter was an important determinant of the number of non-REM dreams collected. That is, the reports made by subjects woken from non-REM sleep were ambiguous and open to interpretation as either thinking or dreaming, while reports from REM sleep awakenings tended to be indubitably of dreams.

What is clear is that mental life does not cease during non-REM sleep, and REM sleep is not a necessary condition for dreaming. Primary process thinking, dream 'logic' and the construction of generally illogical stories all seem to take place throughout the night's sleep. What sets the experience of REM sleep apart is the vividness of the visual imagery and the generally more compelling quality of the dreams associated with it.