Notes on Dreaming: Dreaming as Cognitive Process1

In previous considerations of contemporary dream research, we have surveyed the basic experimental evidence derived from the study of dreaming activity in animals and man and have explored some of the implications of the available evidence for psychoanalytic theory (Meissner, 1968c). In the present paper, our intention is to extend this analysis to the specifically cognitive aspects of the dreaming process. Presumably an examination of these relationships can be expected to shed some further light on both the dream process and cognitive functions.

It has become apparent from more recent studies of mental processes during sleep that there is more than one kind of dreaming (Meissner, 1968c; (Fisher, 1965). After the original studies of dream experience, in which very little dream recall was observed in relation to NREM periods and significant recall was associated with REM periods, it was thought that REM activity and dreaming activity were coterminous if not synonymous (Dement and Kleitman, 1957a), (1957b). Further study, however, made it clear that dreaming also occurs in NREM periods (Rechtschaffen et al., 1963); (Foulkes, 1962). NREM dream activity tends to be more abstract, logical, more plausible, more like thinking, less emotional and more concerned with contemporary waking experience. It has a more reality-oriented, secondary process quality by comparison with the disorganized, emotional and highly visual productions of REM dreaming (Foulkes et al., 1962).

The contrasts in style of thought organization are paralleled by quite different and characteristic patterns of neurophysiological activation. It is to this relationship that we would like to direct our attention for the moment. The REM pattern of organization is characterized by desynchronization in the diencephalic and mesencephalic reticular formation, strikingly synchronous theta-activity in hippocampus, and cortical desynchronization. NREM states show a marked co