Dream mechanisms: Is REM sleep indispensable for dreaming?

Abstract

This article aims to overview briefly major arguments about dream generation from the point of view of the pros and cons of the main thesis that rapid eye movement (REM) sleep is indispensable for dreaming. The methodological issues observed in empirical dream research are also discussed. The evidence, as seen in phenomenological corroboration and anatomical and neurocognitive findings specific to REM sleep, appears to support the contribution of REM mechanisms to dream experiences. As a representative model, Hobson’s “AIM model”, which proposes a three-dimensional state space in mind, is introduced. Evidence for the counter-argument is provided by the appearance of dreamlike mentation outside of REM sleep, psycholinguistic studies of report length, and neuropsychological corroboration between dream cessation and forebrain lesions. As representative models, Solms’ “forebrain dopaminergic mechanisms”, inductively developed by the body of neuropsychological findings, and “the functional state-shift hypothesis”, hypothesizing dream appearance depending on the proximity of electroencephalographic activities to wakefulness, are introduced. As an alternative between the two standpoints, the “covert-REM model” proposes REM mechanisms underlying dreamlike activities outside REM sleep in a covert way. This model has been elaborated through evidence such as the appearance of physiological activities specific to REM sleep preceding the REM period, and an increase in dream recall as its temporal proximity to peak REM activity. A possible contribution of arousal processes to mentation occurrences outside of REM sleep is also proposed, as seen in evidence such as the increase of mentation during nocturnal sleep, with frequent arousals and stimuli-induced mentation during NREM sleep. Other issues to be considered in dream mechanism studies are also discussed.