Llewellyn makes a convincing case that rapid eye movement (REM) sleep serves the function of elaborative encoding by following ancient art of memory (AAOM) rules. At the phenomenological level, this becomes manifest in visualizations, bizarre connections, and narrative form during dreams. Llewellyn summarizes research showing that, during REM sleep, the prefrontal areas are in a state of deactivation resulting in fluid reasoning and flexible thought. However, she touches only briefly upon the ramifications of her analysis for understanding the development and perseverance of psychopathological symptoms.

We would like to emphasize the relevance of Llewellyn's analysis for one particular type of psychopathology – namely, dissociative symptoms (e.g., derealization, amnesia, and absorption). Dissociative symptoms are common in the healthy population, but disorders such as dissociative identity disorder (DID; formerly known as multiple personality disorder) and depersonalization disorder, represent severe, albeit rare, manifestations of psychopathology (Lynn et al. Reference Lynn, Lilienfeld, Merckelbach, Giesbrecht and Van der Kloet2012). The traditional view is that dissociative symptoms reflect (transient) disruptions in memory, perception, and/or consciousness and that these disruptions are causally related to aversive life events (e.g., Dalenberg et al. Reference Dalenberg, Brand, Gleaves, Dorahy, Loewenstein, Cardeña, Frewen, Carlson and Spiegel2012). More specifically, the idea is that dissociative symptoms enable individuals to distance themselves from the emotional impact of aversive events. Although there is some indirect evidence – largely correlational – for this trauma hypothesis, it suffers from one important weakness: it remains silent as to how aversive events produce dissociative symptoms.

More recently, researchers have proposed that sleep disturbances play an important role in the development of dissociative symptoms. A solid and steadily accumulating foundation of research now exists to contend that dissociative symptoms are associated with a labile sleep–wake cycle in which dreamlike mentation invades the waking state, produces memory failures, and fuels dissociative experiences (Koffel & Watson Reference Koffel and Watson2009; Van der Kloet et al. Reference Van der Kloet, Giesbrecht, Lynn, Merckelbach and de Zutter2012a; Reference Van der Kloet, Merckelbach, Giesbrecht and Lynn2012b; Watson Reference Watson2001).

The idea that sleep disturbances and dissociative symptoms are related is not new. In the nineteenth century, double consciousness, the historical precursor of DID, was often described as somnambulism, which refers to a state of sleepwalking. Patients suffering from this disorder were referred to as somnambules (Hacking Reference Hacking1995), and many nineteenth-century scholars believed that these patients were switching between a “normal state” and a “somnambulistic state.” In 2001, Watson investigated two large samples of undergraduate students and showed that dissociative symptoms are linked to self-reports of vivid dreams, nightmares, recurrent dreams, hypnopompic imagery, and other unusual sleep phenomena. His finding has been reproduced time and again. We (Van der Kloet et al. Reference Van der Kloet, Giesbrecht, Lynn, Merckelbach and de Zutter2012a) summarized the findings of 23 studies and found an average correlation of r = 0.41 between dissociative symptoms (as measured by the Dissociative Experiences Scale [Bernstein & Putnam Reference Bernstein and Putnam1986]) and unusual sleep experiences (collected with measures such as the Iowa Sleep Experiences Survey [Watson Reference Watson2001]).

The connection between sleep and dissociative symptoms seems specific in the sense that unusual sleep phenomena that are difficult to control, including nightmares and waking dreams, are related to dissociative symptoms, but lucid dreaming – dreams that are controllable – are only weakly related to dissociative symptoms. Germane to this specificity issue is the study by Koffel and Watson (Reference Koffel and Watson2009) in which 374 participants completed a comprehensive test battery, including measures of psychopathology and sleep. The authors concluded that “unusual sleep experiences are specific to dissociation and schizotypy, whereas insomnia and lassitude are specific to depression and anxiety” (p. 551).

However, these studies on sleep and dissociation used a correlational approach, which precludes the ability to draw causal conclusions. If dissociative symptoms are, indeed, fueled by a labile sleep–wake cycle, sleep loss would be expected to intensify dissociative symptoms, thereby suggesting a specific temporal pattern. We tested this prediction in a pilot study (Giesbrecht et al. Reference Giesbrecht, Smeets, Leppink, Jelicic and Merckelbach2007) that tracked dissociative symptoms in 25 healthy volunteers during one night of sleep deprivation. We found that sleepiness, as well as spontaneous and induced dissociative symptoms, were stable during the first day, but substantially increased after one night of sleep loss. Interestingly, the increase in dissociative symptomatology was highly specific: Dissociative symptoms were affected by sleep loss sooner than were mood deterioration.

The reverse appears to be true, as well. We (Van der Kloet et al. Reference Van der Kloet, Giesbrecht, Lynn, Merckelbach and de Zutter2012a) conducted a longitudinal study to investigate the relation between unusual sleep experiences and dissociation in a mixed inpatient sample (N = 195) evaluated on arrival and at discharge six to eight weeks later. We found a robust link between unusual sleep experiences and dissociative symptoms and determined that sleep normalization was accompanied by a reduction in dissociative symptoms. The link between dissociation and sleep is likely more differentiated, as we observed that decreases in narcoleptic experiences rather than decreases in insomnia accompanied the reduction in dissociative symptoms.

Finally, in a recent study (Van der Kloet et al. Reference Van der Kloet, Franck, Van Gastel, De Volder, Van Den Eede, Verschuere and Merckelbach2013), we measured dissociative symptoms and EEG sleep parameters in patients (N = 45) suffering from insomnia. We found that it is lengthening of REM sleep that predicts dissociative symptoms. This finding is consistent with the hypothesis that a disturbed sleep–wake cycle, possibly due to aversive life events, produces excessive or out-of-phase REM activity that, in turn, underlies dissociative symptoms. These dissociative symptoms may, in turn, exacerbate or increase vulnerability to sleep disturbances, engendering a vicious cycle that may be ameliorated with interventions that target dissociation, sleep problems, or both.

Thus, one distinct scenario that warrants further investigation is that excessive REM sleep during the night and/or minor REM sleep episodes during the day fuel the type of fluid and hyperassociative cognition that is typical for dissociative disorders. This research perspective might shed new light on the propensity of dissociative individuals to develop false memories. Even more importantly, it might suggest new treatment options for dissociative patients. Whereas Llewellyn focused on the memory-promoting aspects of REM sleep, we have emphasized the pathological potential of excessive REM. It would be exciting to combine these two lines of research.