May 1972

Limited Discriminability of REM and Sleep Onset Reports and Its Psychiatric Implications

Author Affiliations

From the Georgia Mental Institute and the Department of Psychiatry, Emory University, Atlanta.

Arch Gen Psychiatry. 1972;26(5):449-455. doi:10.1001/archpsyc.1972.01750230059012

Recent laboratory failures to find a link between dreaming and schizophrenia have been based on the assumption that regressive dreams are regular and unique properties of REM sleep. Results of the present study contradict that assumption. We show that sleep onset stage 1 (SO) reports and REM reports are frequently indiscriminable in terms of regressivity (composite of bizarre, perceptual, emotional, etc). About 25% of 194 SO reports were regressive enough to be called REM reports by each of five trained judges and about 50% of 63 REM reports were sufficiently nonregressive to be called SO. We conclude that an adequate test of the hypothesis that the physiologic correlates of dreaming are those of schizophrenic mentation requires the use of physiologic correlates of SO regressive mentation.