Exploring the possibility that dreams protect sleep (PDF)

1 Dream Function: Exploring the possibility that dreams protect sleep Refqah Jassiem and Danyal Wainstein Department of Psychology University of Cape Town Supervisor: Mark Solms Word Count: Abstract: 168 Main Body: 9780  2 ABSTRACT A review of current literature shows that a physiological function for dreaming has not yet been empirically established. Recent developments have resulted in the identification of the neural correlates of dreaming; however, these correlates are yet to be linked with a function for dreaming. Based on the literature, it is reasonable to propose that dreams may function to protect sleep by diverting the disruptive physiological arousal events that naturally occur during sleep. To test this hypothesis, a single-case of dream loss due to bilateral ischemic infarction in the occipital lobes was studied using polysomnographic recordings. The quality and quantity of the patient’s sleep was analysed over two nights in the sleep laboratory in order to verify whether her sleep was significantly more disturbed than the established norms. As predicted, results showed that sleep was significantly more disturbed than the norms, specifically with regard to non-REM sleep stages 1 and 2. The ways in which dreaming may contribute to the protection of sleep are discussed in light of this finding. Keywords: dreaming; dream loss; bilateral occipital lesions; disturbed sleep; polysomnograph  3 DREAM FUNCTION: EXPLORING THE POSSIBILITY THAT DREAMS PROTECT SLEEP There are many theories that propose a physiological function for dreams; however, due to methodological difficulties, no dream theory to date has been empirically proven (Solms & Malcolm-Smith, 2009). How do you study objectively the most subjective of human experiences? Due to a lack of scientific investigation in this field, numerous speculative dream theories currently exist. Some theories claim that dreaming serves an adaptive evolutionary function (Revonsuo, 2000), while others believe that it helps to consolidate emotional memories during sleep (Breger, 1967). A few theories even propose that dreaming serves no function at all (Crick, 1983; Hobson, Pace-Schott & Stickgold, 2000); and there are those who believe that dreaming performs the very fundamental function of protecting sleep (Freud, 1953; Solms, 1997, 2000). However, unproven dream theories divide the field of dream science, rather than contributing to its advancement. It is therefore important that these dream theories be empirically validated. This study hopes to contribute towards uncovering a sound physiological function for dreams by empirically testing whether dreams protect sleep. Dreaming as an Epiphenomenon of REM Sleep In the 1950s it was established that the sleep cycle consists of various phases, one of which is rapid eye movement (REM) sleep. Rapid eye movement sleep is characterised by bursts of rapid eye movement, accompanied by specific physiological changes that include a loss of muscle tone (atonia) and cerebral cortical activation similar to that of the waking state (Aserinsky & Kleitman, 1953). Among other things, this discovery acted as an impetus for further dream research, which subsequently showed that dreaming and REM are highly correlated (Aserinksy & Kleitman, 1955; Dement & Kleitman, 1957). As a result, Aserinsky and Kleitman (1953) immediately conflated the two phenomena by stating that dreaming and REM were identical states. Following this, Hobson & McCarley (1977) further deduced that dreams are merely an epiphenomenon of physiological REM processes, and therefore serve no independent function. This explanatory model of dreaming is referred to as the activation-synthesis model, and proposes that the forebrain structures that generate dreams are activated by cholinergic brainstem mechanisms (that generate REM sleep), causing meaningless representations (such  4 as thoughts, feelings and images) to be passively synthesised. In this way, the forebrain makes “the best of a bad job in producing even partially coherent dream imagery from the relatively noisy signals sent up from the brainstem” (Hobson & McCarley, 1977, p. 1347). This view of dreaming dominated sleep and dream research for decades. However, the assumption that dreaming and REM sleep are identical, and that both are generated by a single brainstem mechanism, has since been challenged by contemporary neuropsychological evidence (Bishoff & Basssetti, 2004; Poza, & Martí Massó, 2006; Solms, 1997). The Neural Correlates of Dreaming In order to understand how dreaming and REM sleep are doubly dissociable, it is firstly necessary to understand the neural mechanisms that have been found to be essential to the generation of dreams. Extensive clinico-anatomical lesion studies by Solms (1997) have identified the parieto-temporo-occipital (PTO) junction and the white matter of the ventro- mesial quadrant of the frontal lobes as being crucially involved in the dream state, as damage to either one of these regions results in a total cessation of dreaming. Furthermore, several neuroimaging studies have confirmed that these regions are highly activated during REM sleep (Braun et al., 1997; Dang-Vu et al., 2005; Maquet et al., 1996; Nofzinger et al., 1997). These studies repeatedly illustrated that the dreaming brain involves an especially specific group of forebrain structures; namely, the dopaminergic mesocortical-mesolimbic (MC-ML) system. Dreaming is generated by the Dopaminergic Mesocortical-Mesolimbic System The dopaminergic MC-ML system is defined as the “system [that] is formed by dopamine neurons located in the ventral tegmental area…which project to the nucleus accumbens, prefrontal cortex, septum, amygdale, and hippocampus” (Dahan et al., 2007, p. 1232). Converging lines of evidence indicate that this region is crucial to dream production. Firstly, when the dopaminergic pathway that runs through the MC-ML system is transacted, in the surgical procedure known as modified prefrontal leucotomy, cessation of dreaming occurs (Jus et al., 1973; see Solms, 1997, for review). Secondly, L-Dopa—a drug that specifically stimulates dopamine in this region—has been found to intensify the bizarreness, emotionality and vivacity of dreaming, while the REM cycle remains unaltered (Hartmann et al., 1980). Thirdly, a cessation of dream recall has been associated with Parkinson’s disease, which is usually the result of an insidious depletion of dopamine in various forebrain regions (Sandyk, 1997). Fourthly, it has been reported that there is an increase in dopamine release in  5 humans during REM sleep, within the MC-ML system (Gottesmann, 2004). Finally, recent studies based on the internal measurement of dopamine (through microdialysis and single cell recordings) during the sleep-wake cycle in rats, found a substantial increase in dopamine cell activity and terminal release during REM sleep (Dahan et al., 2007; Lena et al., 2005). In summary, the overall consensus between the clinico-anatomical studies on the one hand, and the neuroimaging and in vivo studies on the other, suggest that the dopaminergic MC-ML system is responsible for the neurogenesis of dreams. REM Sleep and Dreaming are Doubly Dissociable States As previously mentioned, there is a substantial amount of evidence in support of the fact that dreaming and REM sleep are doubly dissociable states. A number of case studies provide evidence for dream cessation occurring while the REM cycle has remained entirely unaltered (Benson & Greenberg, 1969; Bishof & Bassetti, 2004; Brown, 1972; Jus et al., 1973, Poza & Martí Massó, 2006). In other words, the patients ceased to dream while still experiencing entirely normal REM sleep states. Moreover, patients who have suffered lesions to their pontine brainstems, resulting in the elimination of their REM sleep cycles, have still reported experiencing dreams (Solms, 1997). Despite such evidence, many researchers dispute the fact that dreaming and REM actually are doubly dissociable states, because these two phenomena frequently co-occur. In response to this, Solms (2000) explains that although dreaming is highly correlated with REM sleep, this correlation is not due to cholinergic brainstem mechanisms generating dreams, but rather because cholinergic activation provides sufficient cerebral stimulation during sleep for dreaming to occur — dreaming can result from any such cerebral activation. To illustrate this point, Solms (1997, 2000a) draws attention to the occurrence of nightmares during non-REM (NREM) sleep in patients experiencing partial seizures. Furthermore, there is a substantial amount of literature documenting dreaming outside of REM sleep, when cholinergic brainstem mechanisms are silent. Furthermore, numerous studies have documented that some form of dream reports can be obtained from all stages of NREM sleep at rates of up to 50-75% (Foulkes, 1962; Foulkes & Vogel, 1965; Suzuki et al., 2004). Moreover, dream reports from NREM sleep that are indistinguishable from those during REM, have been reported between 5-10% (Hobson, 1988) and 10-30% (Monroe et al., 1965, as cited in Rechtschaffen, 1973) of the time. In particular, Foulkes and Vogel (1965) reported that participants awoken at sleep onset and during N1 sleep often report vivid  6 “dream-like” hallucinatory images with kinaesthetic features, as well as static visual imagery throughout the NREM sleep stages. Recently, functional neuroimaging results have provided further evidence for dream activity during NREM. For example, Hofle et al. (1997) used positron emission tomography (PET) imaging techniques to investigate the changes in regional cerebral blood flow (rCBF) during the progression from relaxed wakefulness through slow wave sleep (SWS). The changes were examined as a function of spindle and delta EEG activities that progressed during NREM sleep. Hofle et al. reported that delta activity and rCBF covaried positively in the primary and secondary visual cortex and the secondary auditory cortex (BA 22). Indeed, this activation is similar to the activation recorded when subjects are asked to lay with their eyes closed and imagine sounds or images (Hofle et al., 1997). In addition, positive covariation between delta activity and rCBF was also found in the left inferior parietal lobule (BA 40). In light of their results, Hofle et al. (1997) hypothesised that the visual and secondary auditory cortex may reveal a possible substrate for dream-like mentation during NREM sleep. These results complement the evidence reviewed above, as the activation of the left inferior parietal lobule corresponds with Solms’s (1997, 2000a, 2000b) clinico- anatomical lesion findings. Furthermore, a functional magnetic imaging (fMRI) study by Portas et al. (2000), that examined auditory processing across the sleep-wake cycle, found that neutral auditory stimuli presented during sleep resulted in activation of the auditory cortex, thalamus and caudate bilaterally; while a meaningful auditory stimulus (e.g. the participant’s name) additionally resulted in higher activation of the middle temporal gyrus and the orbitofrontal cortex bilaterally. As a result, these authors concluded that the brain was able to facilitate external stimuli during sleep. Therefore, it seems that the dreaming mind may retain subliminal contact with reality through the sensory channels, suggesting that “sensory stimuli that reach us during sleep may very well become the sources of dreams” (Freud, 1900/1953, p.23, as cited in Solms, 1997, p.136). Unfortunately, neither of these studies actually investigated whether dreaming was definitely related to the neuronal activation seen during the neuroimaging scans. However, Takeuchi et al. (2001), in a study aimed at exploring the relationship between dreaming during induced Sleep Onset REM Periods (SOREMPs) and regular N1 sleep onset, found that dreaming was more likely to occur in NREM sleep onset when there  7 was increased electroencephalographic (EEG) arousals1 and waking during this time. As a result, the authors concluded that there is a “strong relationship between NREMP Dreams and awakening in our results...we postulate that arousal processes might be related to Dream production during NREM sleep” (p. 50). Therefore, Solms’s (1997) hypothesis, that dreaming throughout the sleep-wake cycle can be activated by various arousing stimuli, some of which are external, is supported by numerous lines of evidence. Consequently, many now agree that dreaming is neither intrinsic to, nor isomorphic with, the REM state (Feinberg, 2000; Vogel, 2000). It would appear that dreaming serves a specific function during sleep, and that it is not merely an epiphenomenon of REM processes. Dreams Protect Sleep As mentioned, the dopaminergic MC-ML system is responsible for generating dreams, and is a partial constituent of what Panksepp (1998) refers to as the SEEKING system. This system motivates nonspecific appetitive behaviours within all mammals, including humans. Several converging lines of evidence have already confirmed that this system is highly active during REM sleep, even more so even than during waking (Dahan et al., 2007; Gottesmann et al., 2004). This evidence has led Solms (1997, 2000a) to hypothesise that because the dorsolateral prefrontal cortex (DLPFC) is deactivated during sleep, the volitional urges which are usually executed in this region during waking (in the form of thoughts and actions with logical consistency, structure and volition) have to be redirected. Therefore, Solms (2000a) argues that these appetitive urges eminating from the highly aroused limbic system are regressively directed toward the PTO region2, where they are represented virtually as dreams. 1 Arousals can be defined as transient phenomena that result in fragmented sleep without behavioural waking (p. 10). Specifically, an arousal can be scored during REM, N1, N2, or N3 if there is an abrupt shift in the EEG frequency that is characterised by a 3 to 14 second intrusion of alpha, beta, or theta waves (but not spindles or delta). Arousals are expressed as a number per hour (Arousal Index; AI) and in middle aged adults it is normal to have an AI of up to 10 (Chokroverty, 2009). 2 The PTO junction forms an association cortex in the brain; this region is not responsible for primary sensory experiences, but participates in sensory integration and abstract thought processing (Yu, 2007). As mentioned, this region is crucial to dreaming, and damage can result in complete dream loss (Bishoff & Bassetti, 2004; Poza, & Martí Massó, 2006; Solms, 1997). Therefore, it is reasonable to assert that the PTO junction is the perceptual stage upon where dreams are ‘played out’.  8 Freud (1900) referred to this process as “regression” and argued that “in dreams the fabric of thought is resolved into its raw material” (as cited in Solms & Turnbull, 2002, p. 211). In addition, given the facts that EEG arousals arise from internal and external sources throughout sleep, and that they have been shown to positively correlate with dreaming at sleep onset, we propose that dreaming may also protect sleep by actively diverting these naturally occurring arousals. To date, no study has investigated the relationship between dream loss and EEG arousals that occur throughout sleep. However, we are not the first to propose this function of dreams. Freud’s Dream Theory. In 1900, Freud proposed that dreams “serve the purpose of prolonging sleep instead of waking up. Dreams are the guardians of sleep, and not its disturbers” (Freud, 1953, p. 223). Freud (1953) argued that dreams are part of a process of wish-fulfilment that enables the sleeper to continue sleeping because “the internal demand which was striving to occupy him has been replaced by an external experience, whose demand has been disposed of” ( p. 223). The similarities between Freud’s dream theory, and the hypothesis that dreams protect sleep by acting as an outlet for neurological arousal during sleep, are striking. Moreover, Freud’s dream theory remains a cornerstone of psychoanalysis. Consequently, there is great significance in empirically testing whether or not dreams protect sleep; firstly, in seeking to establish a physiological function for dreaming; and secondly, to challenge Popper’s (1963) conclusion that psychoanalysis is a pseudo-science that is incapable of generating falsifiable hypotheses. Preliminary Findings In a clinical investigation, Solms (1997) found evidence to suggest that a total loss of dreaming subjectively disrupts sleep: of 101 patients experiencing complete dream loss due to numerous brain injuries and illness affecting both the anterior frontal regions, as well as the posterior PTO region, a significant proportion claimed that their sleep had been disrupted since the onset of their injury. In contrast, the 260 controls with equivalent injuries who continued to dream had significantly less disturbed sleep. As a result, Solms (1997) concluded that “pending the outcome of objective (sleep laboratory) studies, which may or may not confirm the subjective reports, the sleep-protection theory of dreams is provisionally supported” (p. 165). However, these results were attained using clinical-bedside interviews  9 and are thus merely suggestive; further rigorous testing, using objective polysomnographic (PSG) data, needs to be completed in order to empirically test this hypothesis. Furthermore, in 2004, Bishof and Bassetti were the first to report a case of complete dream cessation with full clinical, neuropsychological, neuroimaging, and polysomnographic documentation. Their patient, a 73-year-old woman, had suffered bilateral occipital-lobe damage after having a stroke (an acute ischemic infarction due to atrial fibrillation). The patient experienced total dream loss, and began to regain some of her dream recall after 14 weeks. Although the purpose of their study was to show that their patient retained her REM cycle while experiencing complete dream loss, the PSG recording incidentally revealed that the woman had sleep maintenance insomnia.3 In addition, Poza and Martí Massó (2006) recently reported the case of a man who ceased dreaming after a unilateral left temporo-occipital hematoma. The authors documented their patient’s lesions and sleep patterns using neuroimaging and PSG. They reported that the man continued to have intact REM sleep cycles, but that after his brain injury he started complaining about his bad quality of sleep. Therefore, two well documented cases that have reported disturbed sleep with a loss of dreaming exist in the literature to date. However, the authors did not consider the possibility that the disturbances in their patients’ sleep could be a result of their cessation in dreaming. 3 Sleep-maintenance insomnia is the “disturbance in maintaining sleep once achieved; persistently interrupted sleep, without difficulty falling asleep.” (Chokroverty, 2009)  10 Aims and Hypotheses While there has been an increased interest in the scientific exploration of dreams in the last few decades, no dream theory has yet been successfully empirically tested. Furthermore, while there have been two cases of total dream loss published (Bischoff & Bassetti, 2004; Poza & Martí Massó, 2006) with full clinical, neuropsychological, neuroimaging and polysomnographic data, the aim of these studies was to fully document a cessation of dreaming simultaneously occurring with regular REM cycles and not to report the effects of dream loss on sleep. However, both cases reported that their patients had experienced disturbed sleep since their dream loss. In light of these results, this study had two primary aims: firstly, we aimed to use polysomnographic recording to comprehensively document dream loss as a distinct neuropsychological dysfunction; and secondly, to focus on the relationship between disturbed sleep and loss of dreaming. For this study, it was predicted that a patient with bilateral occipital lesions would have complete dream loss and that her sleep would subsequently be disturbed. The following hypothesis was examined: H1: A single participant with bilateral occipital lesions will experience significantly disturbed sleep when compared with published norms.  11 METHODS Sample Mrs P was a 65-year-old dextral female. She was a homemaker with 10 years of formal education, and no previous medical or psychiatric history. The selection criteria for the case included loss of dreaming (a complete lack of subjective dream recall), accompanied by specific posterior lesions to the PTO junction. Mrs P was the first patient found to meet these criteria. Any patients with any other sleep or neurological disorder that could confound the results, were not considered for the study. Mrs P was referred by a neurologist from Groote Schuur Hospital. Healthy Age-Matched Controls. In order to understand specifically what effects dream loss may have on sleep, it is important to exclude natural changes in sleep that are related to age. Age has been found to have disruptive effects on sleep organisation and architecture (Chokroverty, 2009; Feinberg, 1973). For this reason, a study by Boselli et al. (1998) comprised of 10 participants (5 females, 5 males) over 60 years of age, was used as a control group. Boselli et al. recruited healthy individuals with no daytime complaints, good sleep quality and regular life habits. Moreover, the participants’ sleep macrostructure was compatible with widely accepted quantitative norms (Ohayon et al., 2004). Chronic Stroke Controls. In addition to age, stroke has been found to have an impact on the quality of sleep (Bassetti & Aldrich, 2001; Chokroverty & Montagna, 2009; Korner et al., 1986; Vock et al., 2002). In order to exclude the effects of stroke on Mrs P’s sleep quality and quantity, published sleep macrostructure means for a group of patients with ischemic hemispheric lesions have been used for comparison (Vock et al., 2002). Since our case was in the post-chronic phase after stroke (>5 years), Vock et al.’s study was suitable to use for comparison. Measures Case History Mrs P’s case history was taken directly from her medical records, and her case information has been duplicated here in accordance with the APA guidelines for confidentiality and anonymity (American Psychiatric Association, 2005). As such, certain  12 identifying information has been excluded, albeit not to the extent that the information provided has been distorted in any way Dream Recall Mrs P was asked to give a subjective account of her dreaming, or lack thereof, since her stroke in 2006. Specifically, she was asked whether she still dreamt or if she remembered any dreams she may have had in the past 5 years. In addition, semi-spontaneous nocturnal REM sleep interviewing, during the first night in the sleep laboratory, was used to verify dream loss. This interview consisted of briefly asking Mrs P, during unprompted REM awakening, if she was dreaming, and about what was going through her mind prior to being awoken. Neuropsychological Tests A range of neurocognitive tests were chosen for this study, focusing primarily on higher visual and spatial perception, visual and verbal short-term memory, and visual and audio-verbal long-term memory — the necessary neuropsychological functions required for intact dream recall (Appendix A). The constellation of neuropsychological subtests and scoring systems used in this study are widely recognised and internationally established standard measures, and are being used on an ongoing basis in the daily clinical neurocognitive assessments of the neuropsychologists at Groote Schuur Hospital (Strauss, Sherman & Spreen, 2006). Visuo-spatial perception. For the assessment of visuo-spatial perception, the subtests chosen were from Luria’s Neuropsychological Investigation for higher visual perception and integration included: 1) object recognition; 2) visual recognition of letters, words and phrases; 3) calculations; 4) colours and faces; 5) language (Christensen, 1974). In addition, the Judgement of Line Orientation Test (Benton et al., 1994); Benton’s Facial Recognition Test (Benton et al., 1994); and the Boston Naming Test (BNT; which doubles as a language test), were also used (Kaplan, Goodglass & Weintraub, 2001). Constructional praxis. The WAIS-III Blocks were chosen for assessing perceptual organization and constructional praxis (The Psychological Corporation, 1997). In addition, the Rey-Osterrieth Complex Figure was also utilised for this purpose (ROCF; Rey, 1941; Osterrieth, 1944).  13 Short-term memory. Corsi’s Blocks were used to assess visual short-term memory, whilst the Digit Span Test was used to assess audio-verbal short-term memory (WMS-III; Wechsler, 1997). Additionally, the Visual Reproduction-1 test was used to assess visual short memory (WMS-III; Wechsler, 1997). Visual and verbal memory. The ROCF was also used to assess both immediate visual memory and long-term visual memory. The ROCF scoring system includes a copy trial, an immediate recall trial, and a delayed recall trial after approximately 30 minutes. Benton’s Visual Retention was also used to assess long-term visual memory (Sivan, 1992). The Babcock Story was used for the assessment of long-term verbal memory (Babcock & Levy, 1930). In addition, The Bicycle Drawing Test (BDT), the South African flag, and a canary, were all included for the purpose of assessing the patient’s ability to revisualise from memory, without the aid of a copy (Lezak, 1995). Polysomnographic Measures The PSG recordings were completed on a portable Alice © 5 Respironics polygraphic amplifier (Cape Sleep Centre, Gatesville Medical Centre, Cape Town). The American Association of Sleep Medicine (AASM) recommended recording montage was utilized in this study and included: electroencephalogram (EEG; 4 leads, 2 channels); electrooculogram (EOG; 2 channels); the submental electromyogram (EMG; chin and leg); as well as chest and abdominal strain gauges, snore microphone, positional marking and finger pulse oximetry. Sleep stages were visually scored for 30-s epochs by a certified polysomnographic technologist based on AASM standard criteria (Hirshkowitz & Sharafkhaneh, 2009). Sleep macro and microstructure. The analysis of conventional sleep parameters included the total duration of the sleep stages REM and NREM (N1, N2, N3); sleep latency (SL): the interval between lights-out and the first appearance of N1 sleep that subsequently progresses to N2; REM latency (REML): the time it took to reach the first REM stage; total sleep time (TST): total time spent in N1, N2, N3 and REM; total sleep episode (TSE): the total time from sleep onset to end of sleep; wake after sleep onset (WASO): the total time spent awake from sleep onset to the end of sleep; spontaneous arousals: abrupt changes in EEG frequency that may include theta, alpha and frequencies greater than 16Hz, but not sleep spindles. In addition, each arousal must be preceded by at least 10 seconds of continuous sleep, and at least 10 seconds of intervening sleep to score a second arousal (Appendix B).  14 Design An explanatory quantitative single-case study design was used to investigate whether dreams protect sleep. Mrs P’s medical records were used to document the events surrounding her stroke in 2006 and her subsequent dream loss, as well as to verify her lesion localisation. The dependant variables measured in this study were quality of sleep and quantity of sleep. Specifically, quality of sleep refers to sleep efficiency, including whether sleep was fragmented; quantity of sleep refers to sleep architecture, including the time spent in each sleep stage and sleep latencies. Data were collected from full-night attended laboratory PSG recording. The first night was used as an adaption night to control for confounding affects due to the unfamiliarity of the sleep laboratory. The second night was used to measure the quality and quantity for Mrs P’s sleep. The following control measures were also taken: neuropsychological testing was used to assess Mrs P’s memory and visuo-spatial ability, to ensure that she did not have a neurocognitive deficit that might account for her lack of subjective dream recall. Moreover, assessing these functions could potentially help to establish a diagnostic consistency between clinical observation and anatomical pathology. In addition, the effects of age and neurological damage caused by the stroke (apart from dream loss) were controlled for by comparing Mrs P’s PSG sleep parameters with that of normative controls from the literature. The proposed study was nested within a larger study that adhered to the ethical guidelines for research with human subjects as specified by the Health Profession Council of South Africa (HPCSA), as well as the University of Cape Town (UCT) Codes for Research. Ethical approval was obtained from the Psychology Department’s Research Ethics Committee at UCT, as well the Faculty of Health Sciences Research Ethics Committee at UCT (REC. REF. 163/2010). Data Analysis The PSG data was manually analysed and scored by the sleep technologist at Gatesville Medical Centre sleep laboratory, according to the AASM standard sleep guidelines. Thereafter, it was compiled into a comprehensive PSG sleep report using Alice© 5 Respironics software. Statistical Analysis. Hypothesis testing was used to determine whether Mrs P’s quality and quantity of sleep on the second sleep laboratory night was significantly different to healthy age-matched controls (Appendix C). Z-tests were used to transform the Mrs P’s mean  15 scores into z-scores (Zcalc) which were then compared to the critical z-values (Zcrit) in order to determine whether the null hypotheses were rejected. Specifically, the critical z-values used were: Zcrit = ± 2.58 (� = 0.01; [two-tailed test]); Zcrit = 2.32 (� = 0.01, [one-tailed test]; Durrheim, 2002). Neurocognitive Tests. The neuropsychological tests were analysed by way of the hypothetical-deductive approach, as used by the neuropsychologist at Groote Schuur Hospital in their everyday clinical practice. Specifically, we used Mrs P’s test scores to confirm or reject informal hypothesis about her neurocognitive functioning. For example, we hypothesized that Mrs P’s memory would be intact, and used the scores from her tests to either accept or reject this conclusion. Scoring of the neuropsychological tests was done according to the standard procedures outlined with each test. Procedure The neuropsychological testing took place at Groote Schuur Hospital, where a quiet room, free of distractions, was used as an assessment setting. The sleep study was completed at the Cape Sleep Centre at Gatesville Medical Centre—an approved AASM sleep Laboratory — where the PSG recording was professionally monitored by a qualified sleep technologist. Sleep Study Mrs P was monitored in the Gatesville sleep laboratory for two consecutive nights. On both nights full-night PSG recordings from her were obtained. The first night was an adaptation night, to help Mrs P become familiar with the laboratory setting, as well as to confirm her basic sleep/dream activity. The second night Mrs P was left to sleep without interruption in order to determine the quality and quantity of her sleep. Mrs P was thoroughly informed of the main purpose of the study and the procedures, and a consent form was signed before data collection began (Appendix D). Furthermore, she was told that she was free to withdraw from the study at any time without consequence, should she wish to do so. Night 1. Mrs P arrived at the Gatesville Medical Centre at approximately 22:00; she arrived late because she had mistakenly thought that the sleep study was being done at Groote Schuur Hospital, and had gone there first. After being asked to lie down in the sleep laboratory bed, electrodes were attached to her as per the 10-20 system of placement (see