Lucid Dreaming Allows Voluntary Perception During REM Sleep: I. Background and Behavioral Evidence  Stephen LaBerge , Benjamin Baird (LUCIDITY, Tucson, AZ )   C21

Our previous research (LaBerge etal. 1981, 1982, 1986; Brylowski etal 1990) had shown that lucid dreaming occurs in a paradoxically deepened form of REM sleep: that the defining characteristics of PS, both tonic (atonia) and phasic (ANS activation and variability, REMs, etc.) showed significantly elevated measures for LD-REM compared to REM without lucidity (NLD-REM). In particular, H-reflex, a measure of spinal motor inhibition suggested that lucid dreams, rather than being a near-waking, showed a higher degree of suppression of motor output. What about sensory input? It seemed likely that lucid dreaming would show a correspondingly deeper suppression of sensory input. Together, the enhanced disconnection would allow the full cognitive function and world-simulation noted in lucid dreaming, without causing awakening. Thus, we sought to quantify the degree of sensory disconnection using an odd-ball/P300 ERP paradigm during waking and during SVLDs and non-lucid REM sleep. Throughout the night, a pseudo-random sequence of one of two tone pips were presented to the left ear every 1.5-2 sec. Participant-LDers were to respond to any target pips (the high-pitched 1 in 5 auditory stimuli) that happened to be incorporated into their LDs with eye-movement signals. Remarkably and unexpectedly, when participants became lucid, they found that they were able to attend to the auditory stream, and consciously perceive, discriminate, and correctly respond to the targets while otherwise continuing to dream, during unequivocal REM sleep. The broad implication is that attention is the primary determinant of whether or not we consciously perceive environmental events during REM dreaming.