Given its robust and revealing features, it is surprising that dream lucidity was not recognized by philosophers of mind until recently (Metzinger 2003, 1993; Noreika et al. 2010; Revonsuo 2006; Windt in press; Windt & Metzinger 2007). Although it was described by Aristotle (without using the term, in 350 BC), lucid dreaming first appears in the experimental literature of the late nineteenth century (Maury 1861; Saint-Denis & Marquis 1982). It was described as a vehicle for self-experimentation (Arnold-Forster 1921) in the early 20th century and reported on subjectively (van Eeden 1969). The modern laboratory study of lucid dreaming was pioneered by Hearne (1978) and LaBerge, beginning in 1980 (1980).
In this paper, we will summarize our five years of scientific research on lucid dreaming, provide a systematic overview of our work, and present new hypotheses about the why (because of fluctuations in brain networking) and the how (through local changes in lower gamma band activity) of lucid dreaming. Regarding the why, our “Brain Maturation Hypothesis” proposes that the probability of lucid dreaming occurring spontaneously is strongly enhanced during the time of cerebral diversification and, most importantly, integration of the frontal lobes into the cortico-cortical and cortico-thalamic networks (Fuster 1989; Goldman-Rakic 1987; Zilles et al. 1988).
As to the how of lucid dreaming, we will outline our experimental findings, focusing on the increase in lower gamma band activity in fronto-temporal brain areas (Gamma Band Hypothesis). We will then move from our first attempts to provide a brain-based explanation of empirical findings (Hybrid State Hypothesis, see Hobson & Voss 2011; Voss et al. 2013) to a three-dimensional model of consciousness, allowing for a more structured classification of various states of consciousness, ranging from near-death to highly vigilant wakefulness (“Space of Consciousness”, SoC, compare Voss & Voss 2014).
The presentation of our empirical contributions will begin with a quantitative analysis of lucid dream subjectivity. This study demonstrates that the most robust difference between lucid and non-lucid dreaming is the increase in insight into the nature of one’s current conscious state that accompanies lucidity. Based on admittedly few recordings (unpublished) of false awakenings, we are currently inclined to assume that there is no notable difference between, for example, the apparent but non-veridical insight accompanying a false awakening and actual (lucid) insight into the fact that one is dreaming. In both falsely and correctly perceived insight into the current state of arousal, the brain apparently operates in a dissociative mode, allowing for a state-related form of meta-awareness similar to the awareness of mind-wandering described for the wake state (Schooler et al. 2011; Metzinger 2013). However, as our experimentally deduced hypotheses are based on those instances in which the dreamer correctly achieved insight into the fact that she was dreaming while the dream continued, we will restrict our discussion of dream lucidity to these instances.
In discussing these results, we will go so far as to suggest that lucidity, as the name implies, is insight. We then turn to sleep laboratory studies revealing that the principal brain correlate of lucid dreaming is 40 Hz activation of the frontal cortex. When we electrically stimulated the frontal brain via the scalp, we were able to induce both an increase in 40 Hz brain activation and the subjective experience of lucidity. In our discussion of these results we suggest that the experimental study of lucid dreaming is a powerful paradigm for understanding the brain basis of conscious experience.