Lucidity is commonly defined as awareness that one is dreaming while one is dreaming (for excellent reviews, see Voss & Hobson 2015; Dresler et al. in press). Often, this is associated with an ability to control not just one’s own actions in the dream, but also the course of the dream, the actions of non-self dream characters, etc. In particular, lucid dreamers can signal that they have now become lucid by making prearranged patterns of eye movements, such as looking right – left – right – left within their dream. These gaze shifts correspond to the movements of their physical eyes and can be identified on the electrooculogram. This technique of signal-verified lucid dreaming enables researchers to identify the precise period of sleep during which certain actions were performed during a lucid dream and potentially to identify their electrophysiological and neural correlates (Dresler et al. 2011, 2012). Lucidity can occur spontaneously, but a number of methods for inducing lucidity are discussed in the literature (Stumbrys et al. 2012). There have even been suggestions and attempts to experimentally induce dream lucidity through electrical stimulation (Noreika et al. 2010a; Voss et al. 2014; Voss & Hobson 2015). While still in its early stages, this work clearly shows that lucidity is a robust phenomenon; and combined with the ability to control the dream as it unfolds, it makes laboratory studies of lucid dreaming compelling.
One reason for being interested in lucid dreams within the present context are reports of lucid dreams describing a loss of phenomenal embodiment, or even a dissolution of the self (see Windt 2015, chap.s 7, 11 for discussion). Some of these appear to fulfill the requirements for minimal phenomenal selfhood described earlier: in so-called imageless lucid dreams (Magallón 1991; Bogzaran 2003; Hurd 2008), self-identification may be relative to a disembodied point in space and can arise independently of bodily sensations and even of visual imagery (see also LaBerge & DeGracia 2000). While most of these reports, so far, are anecdotal, it is tempting to think that lucid dreams could be used to systematically investigate the transition from minimal phenomenal selfhood to more complex forms of self-experience involving the experience of being a thinking self and embodied agent. Importantly, according to some of these reports, even this basic sense of self-identification and location within a larger spatial expanse can be lost. I would like to suggest that such cases may involve a shift from a simple form of lucid dreaming involving minimal phenomenal selfhood to lucid dreamless sleep experience. Here is a single example:
I am suspended in space—dream space, I think. There is nothing here, just millions of greyish dots and I am one of the dots, there’s no dream-body anymore, I’m just a dot [of] pure consciousness suspended. A feeling of great peace comes over me and a sense of gentle, infinite expansion. It’s as if everything and nothing are the same thing and there is a sense of effortless belonging. As the sense of expansion increases I am no longer a single dot of consciousness; all the dots are me and I am them. There’s no “I” or “them.” We are one. There’s just a blissful sense of timelessness and oneness and a merging with the light. After an indefinable length of time, I start to feel the weight of my body in bed, and settle back into it, tingling all over. (Clare Johnson, unpublished dream report, March 19, 1995)
If we take the report at face value, it describes a gradual transition from minimal phenomenal selfhood, characterized by phenomenally disembodied spatiotemporal self-location, to selfless experience. This transition is accompanied by a sense of spatial expansion, in the course of which the sense of the self as distinct from the environment is lost. To the extent that there still is a sense of spatial self-location, this no longer involves the experience of being located relative to something else. There is also a change in the temporal structure of experience, almost as if the experiential present, the phenomenal now, had been stretched indefinitely. The period following the dissolution of the self is still experienced as having duration, but this duration is indefinable and no longer structured around any events.[23] Following Metzinger (2013), we might want to describe this as involving a transition from a minimal unit of identification, in which an unextended point in space is described as the locus of the self, to a maximal unit of identification. In such cases of “pure consciousness”, he suggests, the unit of identification is
the most general phenomenal property available for identification at all: Philosophers might call it the global “unity of consciousness”, or phenomenality per se, or awareness as such, namely the singular, integrated, all-pervading quality of consciousness characterizing the current totality of experiential contents, as it is given in every single moment of experience. (Metzinger 2013, p. 5)
I would like to suggest that we can now be more precise. The moment at which self-location dissolves—or at which minimal phenomenal selfhood is replaced with the maximum unit of identification—involves a transition to the type of pure subjective temporality that earlier, I suggested might be the phenomenal mark of dreamless sleep experience. As lucid dreaming gives way to lucid dreamless sleep experience, minimal phenomenal selfhood shades into pure phenomenality, in which phenomenal experience is characterized only by its temporal structure. I find it telling that according to Johnson’s report, this latter part of the episode appears to strain the limits of reportability, and also that despite its indefiniteness, the experience is described as blissful; again, this is exactly what the Indian focus on the experience of having slept peacefully would lead us to expect.
Clearly, this single dream report presents anecdotal evidence at best; still, I would like to suggest that a first step towards extending the investigation of dreamless sleep experience beyond experienced meditators might be to investigate imageless lucid dreams in experienced lucid dreamers. What makes me cautiously optimistic is that lucidity is often described as a very unstable phenomenon, as involving a balancing act between maintaining lucid insight (rather than slipping back into a nonlucid dream or awakening) and remaining engaged enough in the ongoing dream to prevent it from dissolving completely (Brooks & Vogelsong 2000). Lucid dreamers often describe that imagery can take on a faded, washed out quality, or that lucidity is followed by a period of darkness or, alternatively, of light; indeed, this may be why such reports often slip into mystical language to describe such experiences. Here, I want only to suggest that in such cases, the unwanted fading of lucid dream imagery may actually be an opportunity for experimentally investigating the transition to dreamless sleep experience.[24]
Before moving on, I want to suggest that the comparison between lucid dreaming and lucid dreamless sleep is also interesting for another reason. This is that as is the case for nonlucid dreams, there continue to be a number of conceptual uncertainties about how to define lucid dreaming and whether to describe it as a genuine sleep phenomenon or as a hybrid state between REM sleep and wakefulness (Voss et al. 2009; for a discussion of lucidity and insight from a philosophical perspective, see Kühle 2015; see Voss 2015 for a criticial reply). Also, while some authors consider any dream involving insight into the fact that one is now dreaming as lucid, others reserve the term lucidity for cases in which there is a marked increase in the overall vividness of multimodal imagery as well as a shift towards wake-like cognitive activity, including the ability to engage in rational thought, full recall of waking life, and insight into the fact that none of the events occurring within the dream have any real-world consequences (for a first attempt to test these different conceptions of lucidity experimentally, see Voss et al. 2013).
On the conception that I favor, lucidity is not necessarily accompanied by an all-pervading change in the phenomenal character of the dream; rather, lucid dreams are gradually distinguished from nonlucid ones along a number of dimensions (Windt & Metzinger 2007; Noreika et al. 2010a; Voss et al. 2013). While laboratory studies, because of their reliance on signal-verified lucid dreams, necessarily focus on lucid dreams involving at least some form of control, the conceptually mediated insight into the fact that one is now dreaming is orthogonal to the other experiential qualities of dreaming. Insight is also necessary to score a given report as describing a lucid dream—but aside from this methodological fact, the ability to conceptualize one’s ongoing experience as a dream—to have the thought “I am now dreaming”—can coexist alongside the types of vivid, often bizarre and emotionally charged imagery and erratic reasoning that characterize a majority of nonlucid dreams as well. Lucidity can be the outcome of a conscious inference (of the type “this cannot be happening, so I must be dreaming”), but often appears to be driven by a sudden feeling, sometimes described by saying that the dream suddenly took on a dreamlike feel or a hyperreal character (see Windt 2015, chap. 9 for details and further references). Perhaps, this precursor to full, conceptually mediated lucidity is similar to the type of nonconceptual awareness that is thought to accompany lucid dreamless sleep experience as well. This suggests two further questions. The first is whether nonlucid forms of dreamless sleep experience exist as well. The second is whether in dreamless sleep experience, anything analogous to prelucid dreams exists. I discuss these in turn.
Again, we can approach the project of identifying candidates for nonlucid dreamless sleep experience by asking whether instances of minimal phenomenal selfhood exist in nonlucid dreams. If so, we could once more expect these to occur in the vicinity of minimal phenomenal experience during dreamless sleep.
A possible candidate for such a state are so-called white dream reports, in which subjects describe the impression of having experienced a dream but are unable to describe it in any detail. It seems plausible that a subgroup of white dream reports can be explained by forgetting. Especially where the subject describes the distinct feeling of having had a complex dream but being unable to remember it in any detail, this would seem to be the most plausible interpretation. There is some reason for thinking, however, that this may not be the case for all reports of white dreaming. In at least some cases, the impression of having had some kind of experience prior to awakening, coupled with an inability to describe any particular aspects of the experience, such as any specific forms of imagery or narrative contents, might not be an artifact of forgetting, but might reflect the structure of the experience itself. At least a subset of white dreams might involve a sense of spatiotemporal self-location, or minimal phenomenal selfhood, arising in an otherwise imageless nonlucid dream. And if this were supported by future studies, then it might even make sense to ask whether perhaps, a further subgroup of white dreams could more properly be described as involving nonlucid dreamless sleep experience. In the current framework, these latter types of white dreams would not count as proper dreams at all: they would be instances of pure subjective temporality arising independently of the spatial aspects of self-location and self-identification. They would involve a form of minimal phenomenal experience that could no longer be described as minimal phenomenal selfhood, and thus as a dream. Perhaps, we occasionally really do retain some awareness, after awakening, of phenomenal experience having persisted during sleep. And perhaps, unable to remember any specific details, we then assimilate them to more familiar types of experiences, labeling them as white dreams.
Again, all of this is still extremely speculative and everything I have said so far about white dreams should be read, at best, as a careful prediction of what we might say in light of future findings. In particular, I do not mean to suggest that white dream reports, or a subgroup thereof, can already be regarded as examples of dreamless sleep experience: I only mean to propose that they are an initially promising target for future research on dreamless sleep experience. Still, these considerations fit in nicely with the finding that white dreams are particularly frequent during slow-wave sleep. According to one study, awakenings from stages 2 and 3 NREM sleep were followed by roughly equal rates of dream reports, white dream reports, and reports of nondreaming (Noreika et al. 2009). Their occurrence in the vicinity of reports of dreaming and of nondreaming might indicate that white dream reports describe a transitional state between the two. Moreover, even dream reports obtained following awakenings from these sleep stages were often static, describing experiences lacking narrative progression as well as movement sensations (Noreika et al. 2009, 2010b). Participants sometimes described the sense of being present in a static scene, as in quietly sitting on a bench, with nothing else happening (Valdas Noreika, personal communication; see also Noreika 2014, p. 52). Even in the absence of narrative progression, there was still a sense of duration, and according to subjective estimates, these simple dreams lasted between thirty seconds and one minute.
An interesting possibility could be to investigate the wording of white dream reports in more detail. To my knowledge, this has not yet been done. Maybe there are indeed subtle differences in the wording of such reports, and perhaps these would enable researchers to distinguish cases in which there is the impression of having forgotten a complex dream from ones describing imageless and perhaps even selfless and objectless episodes of phenomenal experience. Again, it might be possible to increase the expressive granularity of reports with the help of training or specific questionnaires, thus rendering subtle phenomenological differences visible that would be otherwise overlooked. A possible finding could be that some of these experiences involve a continued sense of presence and self-location in an abstract, amodally experienced spatial expanse, whereas in others, even this basic sense of self is lost and only the feeling of duration, or of an indefinite temporal expanse, is present.
A particularly promising way to do this would be to use a serial awakenings paradigm, in which participants are awakened multiple times throughout the night at intervals of 15-30 minutes, thus maximizing the number of reports that can be collected throughout the night (Noreika et al. 2009; Siclari et al. 2014; Siclari et al. 2013). Questions focusing on the temporal aspects of experience could then be used to identify those periods, if any, in which dreamless sleep experience is most likely to occur. For instance, Siclari et al. (2013) asked their participants to estimate how long they had been having continuous experiences before being awakened, but also how long their most recent experience had lasted, how far back in time they could recall any narrative events, and how rich and complex the experience was. They found that during stages N2 and N3, estimates for duration, recall back in time and richness were low. Still, these results could be influenced, in part, by the fact that the interview questions focused on the objects of consciousness and on narrative events. If the questions were reworded in such a way as to cover dreamless sleep experience, the patterns of responses might change. Even so, it is interesting to note that during sleep onset, there was a dissociation between these measures, with participants estimating a long duration of the last conscious experience, but a low richness and ability to recall back in time. At least, this suggests that the estimated duration of conscious sleep states does not always map cleanly onto the ability to recall specific contents. For now, I want only to suggest that a similar strategy could interestingly be applied to the investigation of dreamless sleep experience as well.
This is also attractive in view of the goals of this line of research. Note that Siclari et al. (2014) explicitly use the serial awakenings paradigm to contrast the presence and absence of conscious experience independently of task performance and within the same sleep stage (for a similar suggestion, see Noreika et al. 2009; Noreika 2015), the ultimate aim being to identify the task- and state-independent neural correlates of conscious experience. For this project, dreamless sleep experience, as a candidate for minimal phenomenal experience during sleep, is clearly a relevant target phenomenon.
The final example that I wish to discuss is sleep misperception in subjective (or paradoxical, as it is also sometimes called) insomnia. The term objective insomnia, reserved for patients suffering from actual sleep loss as conventionally measured, is sometimes contrasted with subjective insomnia, which refers to subjects who systematically underestimate the time they actually spend asleep (Harvey & Tang 2012; Perlis et al. 1997). This mismatch between subjective sleep perception and objective measures of sleep sometimes leads to a trivialization of subjective insomnia—and the suggestion that their diagnosis as insomniacs is somehow not “real” can be experienced as infuriating by those afflicted by it (Greene 2008). Subjective insomnia is clearly not an imaginary problem, but a cause of real suffering. In fact, patients with subjective insomnia may experience more severe impairments in cognitive functioning than insomnia patients who do not underestimate the amount of sleep they are getting. Furthermore, worrying about getting enough asleep may precede actual sleep loss, and patients who underestimate the time they spend asleep may still be suffering from a real sleep deficit as well (Harvey 2002; Harvey & Tang 2012). The distinction between subjective and objective insomnia has also been questioned, as sleep-state misperception may be prevalent in different subtypes of insomnia. As Harvey & Tang (2012) put it, “many patients with insomnia perceive sleep as wake, systematically overestimate the time they take to get to sleep (SOL) and underestimate the time they sleep in total (TST).” This further highlights the urgency of sleep-state misperception existing alongside actual sleep loss in insomnia.
In the context of the present discussion, the example of sleep-state misperception in subjective insomnia may seem to be a counterexample to, rather than a candidate for, dreamless sleep experience. Thompson (2015, p. 5) considers sleep-state misperception as a possible objection to his view: sleep-state misperception of the type seen in insomnia challenges the reliability of subjective reports of sleep, thus providing a counterexample to his claim that at other times, reports of dreamless sleep experience and of having slept peacefully are veridical memory reports. He then argues that the mere possibility of there being veridical reports of dreamless sleep experience is enough to disprove the default view. He also proposes that in experienced meditators, “we should observe a stronger correlation between subjective reports of phenomenal qualities of sleep and various objective measures of brain activity” (p. 16). The fact that at other times, subjective evaluations of sleep can go wrong does not contradict this view, but merely shows that the investigation of dreamless sleep experience is best restricted to certain subject groups.
Here, I want to suggest that an alternative interpretation of sleep-state misperception is possible. In this alternative view, patients with subjective insomnia are in fact unwitting experts of various kinds of sleep experience. It is merely in conceptualizing their sleep states as occurring in wakefulness that they go wrong. Yet, this is compatible with saying that during sleep, they maintain prereflective awareness of their ongoing sleep state; in fact, it might be their continued perception of sleep that leads them to mischaracterize it as a state of wakefulness, rather than as sleep. Their expertise, consequently, is of a somewhat paradoxical nature: they have a high-degree of familiarity with their sleep, they observe and perhaps even compulsively attend to it—but they don’t recognize or conceptualize it as sleep.
Note that this description fits in well with the distinction, introduced at the end of section 4_1, between different readings of the term lucidity. There, I argued that prereflective awareness of the fact that one is now dreaming often precedes the conceptually mediated insight that characterizes full-blown lucidity. Importantly, these two factors may even be dissociable: a fleeting awareness of the dreamlike nature of one’s current state can be misinterpreted, on the level of conscious, conceptually mediated thought, as indicating that one is awake. In such prelucid dreams, the erroneous conclusion that one is certainly awake may be prompted by the same type of change in experiential character that in other cases drives the cognitive realization that this is a dream (see also Windt 2015, chap. 9).[25]
Similarly, the idea is that sleep-state misperception arises when patients misinterpret mental activity and phenomenal experience that in fact occurs in sleep as occurring in wakefulness. Indeed, Mercer et al. (2002) found that when they were awakened 5 minutes after the onset of stage 2 sleep or REM sleep, insomnia patients were more likely than good sleepers to say they had been awake. One possibility is that these patients generally have a heightened awareness of sleep-related experiences; another is that increased attention to and concern about the amount of sleep they are getting may increase their sensitivity to such sleep-related experiences, as well as the likelihood of misdescribing them as occurring in wakefulness.
Interestingly, it does not seem that subjective insomnia simply results from a general deficit in the ability to estimate time (Tang & Harvey 2005). Instead, subjective insomnia appears to be associated with selective attention to and increased monitoring of external cues (such as the time of day or the alarm clock), but also of thoughts and bodily sensations that are taken, by the subject, to be inconsistent with sleep. As Mercer et al. (2002, p. 565) put it, “insomniacs’ reduced sleep-wake discriminability may be caused by either a greater amount of mentation during sleep, mentation that more closely resembles awake mentation, or a misattribution of normal nocturnal mentation as wakeful cognitive activity.” Enhanced memory processing may also play a role (Perlis et al. 1997), as might enhanced physiological and cortical arousal. Intriguingly, insomnia patients show heightened beta and gamma EEG activity during sleep onset, but also during NREM sleep; and in one study, this activity was negatively associated with their ability to correctly perceive that they had been asleep (Perlis et al. 2001). Again, this could be an indication of continued awareness during sleep. Subjective insomniacs may be witnessing sleep whilst failing, unlike the expert meditators described by Thompson (2015, see especially p. 16), to realize what it is they are witnessing.
As is the case for white dreams, I am not suggesting that sleep-state misperception in insomnia be equated with dreamless experience, or indeed that any simple explanation is available. Clearly, a wide range of conscious mental activity occurring in sleep might be perceived as occurring in wakefulness, and much of this might be quite different from the specific type of dreamless sleep experience I am interested in here. And equally clearly, sleep-state misperception in insomnia is a far cry from the peaceful type of sleep experience describe in the Indian debate. In her book-length treatment of insomnia, in which she synthesizes research findings with her own personal experience of insomnia, Gayle Greene (2008) describes her reaction to being told, after a sleepless night in the sleep laboratory, that she has in fact been asleep:
So that’s why nobody had come in with a sleeping pill—the EEG said I was asleep. But I was not asleep. I was truly awake. What in the world was it recording? I may have been in a state of deep relaxation, semi-meditative, I usually am when I lie there, and I may have dropped off, but I was aware of all those thoughts, the feel and look of the room, the long drawn-out boredom of lying there without a book to listen to—it felt like consciousness to me. How could I be aware of all that if I hadn’t been awake? (Greene 2008, p. 254)
When asked, however, if she had been aware of the technician coming into her room, she was not (Greene 2008, p. 253).
Here, I want only to make room for the idea that a subgroup of instances of sleep-state misperception might be more properly described as resulting from an awareness of what is in fact sleep, but then is erroneously categorized as belonging to wakefulness. And at least a portion of this awareness of sleep might consist of dreamless sleep experience, or the persistence of temporal experience devoid of further intentional content or any specific objects of awareness during sleep. Moreover, this may well be the dreamless-sleep analogue of prelucid dreaming, where heightened awareness of one’s ongoing state leads to its erroneous characterization as wakeful activity on the level of conceptually mediated thought.
Finally, sleep-state misperception of this type may not be unique to insomnia, but may be prevalent in the general population. In a paper aptly titled “The perceptual uncertainty of having sleep”, Sewitch (1984) describes the outcome of an experiment investigating the ability of healthy subjects to correctly say whether they have been asleep, as determined by objective markers such as EEG measures. She found that out of 210 awakenings from Stage 2 sleep, 116 were judged to be periods of wakefulness; for REM sleep awakenings, the number was lower, with 45 out of 165 awakenings being judged to have been preceded by a period of wakefulness. The surprising conclusion is that even ordinary sleepers quite dramatically underestimate the amount of time they have been asleep (see also Webb 1975). Other studies point in a different direction. There is some evidence that whereas insomnia patients underestimate their total sleep time, healthy subjects overestimate how long they have been asleep (Means et al. 2003; Pinto Jr. et al. 2009). Clearly, more research is needed. But either way, it would seem that our confidence in our ability to tell whether and how long we have been asleep or awake is overrated. And perhaps, at least part of this confusion stems from the fact that the default view is deeply engrained not just in cognitive neuroscience, but also in folk-psychology: We expect sleep to be a state of unconsciousness, and so when we recall mental activity that is distinct from dreaming, we mistakenly think we must have been awake.
There is, however, an even deeper conceptual point. The mismatch between subjective and objective (behavioral or polysomnographic) markers of sleep should alert us to the fact that conventional definitions of sleep, and attempts to operationalize them scientifically, for instance in the form of sleep-stage scoring systems, may be oversimplified. The borders between sleep and wakefulness themselves may be fluid. This brings us back to Thompson’s proposal that a more fine-grained and phenomenologically informed taxonomy of sleep states is needed. This is emphatically illustrated by the following quotation from one of the participants in Sewitch’s study. This participant had subjective insomnia and claimed to have been awake following 22 out of 23 Stage 2 NREM sleep awakenings.
Also, there is for me a state which may be technically sleep to you, but is wakefulness to me and, uhh—it' s an intermediate state—it' s very hard to define, uhh—but I definitely felt that it's there—and uhh—uhh none of the questions precisely examined this situation. (Sewitch 1984, p. 257)
As Thompson suggests, dismantling the default view may be as simple as asking the right kinds of questions.
I began this commentary by formulating a number of related challenges to Thompson’s analysis of dreamless sleep experience. The first two of these centered on the status of reports of dreamless sleep experience. In order to place the scientific investigation of dreamless sleep experience on solid methodological grounding, it is not enough to establish the logical possibility of veridical reports of dreamless sleep experience; rather, some rationale for distinguishing veridical reports from nonveridical ones is needed. Also, in order for such reports to be indicative not just of the occurrence of dreamless sleep experience, but also of its distribution and quantity across sleep, one will have to assume such experiences to be reportable. This means that positive experience reports and reports of an absence of experience, when gathered under the same reporting conditions and unless there is any empirical evidence of disturbing factors, will have to be considered as equally trustworthy. I responded to these dual challenges by pointing out that the methodological background assumptions upon which scientific dream research has long relied, at least implicitly, directly speak to both issues: Dream reports, at least when gathered under (sufficiently) ideal reporting conditions, are indeed assumed to be trustworthy sources of evidence with respect to the occurrence and phenomenal character of experience during sleep (I called this the transparency assumption), and dreams are also assumed to be assumed to be reportable experiences (I called this the reportability assumption). Elsewhere (Windt 2013, 2015), I have argued that both assumptions are theoretically justified because they best explain dream reporting behavior. Here, I only defended the more limited claim that scientific dream research already offers the methodological resources to turn the study of dreamless sleep experience into a scientific research program. This shifts the burden of proof: in order to meaningfully challenge the report-based investigation of dreamless sleep experience, the methodological background assumptions of scientific dream research will have to be challenged as well.
An important upshot was that the default view is inconsistent with scientific dream research. Due to its methodological background assumptions, scientific dream research is committed to the view that if experiences fitting the profile of dreamless sleep experience are, at least occasionally and under sufficiently ideal conditions (for instance immediately after awakening), reported to occur in sleep, then dreamless sleep experience exists. The default view, understood as an a priori and conceptually motivated rejection of dreamless sleep experience, is flawed. I then argued that by taking the analogy between contemporary philosophical and scientific work on dream reports and the Indian debate seriously, valuable lessons can be learned in the other direction as well. In particular, Thompson’s reconstruction and critique of the Nyāya syllogism suggests that certain skeptical objections to the trustworthiness of dream reports run into the same problems, resulting either in circularity or an infinite regress.
The second and third parts of my commentary were dedicated to the third challenge to Thompson’s view. This was that even if dreamless sleep experience exists, and even if reports of dreamless sleep experience are taken to reflect this fact, its occurrence in experienced meditators is too remote to warrant the large-scale revision of sleep-state taxonomy proposed by Thompson. I attempted to meet this challenge, first, by first sketching the outlines of a conceptual framework for describing dreamless sleep experience. In this framework, dreamless sleep experience is characterized by pure subjective temporality, or the experience of duration and of an extended presence (a stretched phenomenal now) arising independently of any further intentional contents, objects of awareness, or modality-specific imagery. This model extends existing work on dreams, where I argue that the simplest forms of dreaming are examples of minimal phenomenal selfhood, or self-location in a spatiotemporal reference frame (Windt 2013, 2015). In dreamless sleep experience, even this minimal form of self-experience is lost; pure subjective temporality during dreamless sleep experience is a candidate for minimal phenomenal experience, or the simplest form of phenomenal consciousness.
In the final part of the commentary, I discussed what I take to be the most plausible candidates for dreamless sleep experience in this sense: these are lucid dreamless sleep, white dreams, and sleep-state misperception as most prominently seen in subjective insomnia. I also proposed that these states can be meaningfully compared to the transition from nonlucid to prelucid and fully lucid dreams. Here, my aim was to show that dreamless sleep experience is not a remote possibility, but might plausibly turn out to be a common characteristic of sleep.[26]
Importantly, I am not claiming that the proposed conceptual model is the final word on dreamless sleep experience; it is only a very first attempt to delineate the borders of the target phenomenon. The model is clearly open to further conceptual refinement, and I would like it understood mainly as an invitation to do so. What I would hope, however, is that the model might facilitate this process by guiding and informing future research. Similarly, the empirical candidates for dreamless sleep experience that I propose should not be taken to be exhaustive, and their plausibility will depend on future research findings. For now, I hope, however, that they lend further support and urgency to Thompson’s case for dreamless sleep experience.
Acknowledgments
I would like to thank Roxane Dänner, Martin Dresler, and Valdas Noreika for insightful conversations and helpful comments on an earlier version of this manuscript. And I would like to thank Thomas Metzinger for his guidance, patience, and thoroughly constructive criticism.