Matches in SemOpenAlex for { <https://semopenalex.org/work/W2912551200> ?p ?o ?g. }
- W2912551200 endingPage "264" @default.
- W2912551200 startingPage "253" @default.
- W2912551200 abstract "Free AccessRelationships - Stress - Fear - Depression - Anxiety - Nightmares - Dreams - Behavior - Aging - Scientific InvestigationsNightmare Severity Is Inversely Related to Frontal Brain Activity During Waking State Picture Viewing Louis-Philippe Marquis, BSc, Sarah-Hélène Julien, BSc, Andrée-Ann Baril, PhD, Cloé Blanchette-Carrière, BSc, Tyna Paquette, MSc, Michelle Carr, PhD, Jean-Paul Soucy, MD, Jacques Montplaisir, MD, PhD, Tore Nielsen, PhD Louis-Philippe Marquis, BSc Department of Psychology, Université de Montréal, Montréal, Québec, Canada Center for Advanced Research in Sleep Medicine, CIUSSS-NÎM – Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada Search for more papers by this author , Sarah-Hélène Julien, BSc Department of Psychology, Université de Montréal, Montréal, Québec, Canada Center for Advanced Research in Sleep Medicine, CIUSSS-NÎM – Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada Search for more papers by this author , Andrée-Ann Baril, PhD Center for Advanced Research in Sleep Medicine, CIUSSS-NÎM – Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada Department of Psychiatry, Université de Montréal, Montréal, Québec, Canada Search for more papers by this author , Cloé Blanchette-Carrière, BSc Center for Advanced Research in Sleep Medicine, CIUSSS-NÎM – Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada Department of Psychiatry, Université de Montréal, Montréal, Québec, Canada Search for more papers by this author , Tyna Paquette, MSc Center for Advanced Research in Sleep Medicine, CIUSSS-NÎM – Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada Search for more papers by this author , Michelle Carr, PhD Sleep Laboratory, Swansea University, Swansea, United Kingdom Search for more papers by this author , Jean-Paul Soucy, MD Montreal Neurological Institute, Montréal, Québec, Canada Search for more papers by this author , Jacques Montplaisir, MD, PhD Center for Advanced Research in Sleep Medicine, CIUSSS-NÎM – Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada Department of Psychiatry, Université de Montréal, Montréal, Québec, Canada Search for more papers by this author , Tore Nielsen, PhD Address correspondence to: Tore Nielsen, PhD, Dream and Nightmare Laboratory, Center for Advanced Research in Sleep Medicine, CIUSSS-NÎM – Hôpital du Sacré-Coeur de Montréal, 5400 Gouin Blvd West, Montreal, Que., Canada H4J 1C5+1 514 338 2222x3350+1 514 338 2531 E-mail Address: [email protected] Center for Advanced Research in Sleep Medicine, CIUSSS-NÎM – Hôpital du Sacré-Coeur de Montréal, Montréal, Québec, Canada Department of Psychiatry, Université de Montréal, Montréal, Québec, Canada Search for more papers by this author Published Online:February 15, 2019https://doi.org/10.5664/jcsm.7628Cited by:8SectionsAbstractPDFSupplemental Material ShareShare onFacebookTwitterLinkedInRedditEmail ToolsAdd to favoritesDownload CitationsTrack Citations AboutABSTRACTStudy Objectives:Growing evidence suggests that nightmares have considerable adverse effects on waking behavior, possibly by increasing post-sleep negative emotions. Dysphoric reactions to nightmares are one component of nightmare severity for which the neural correlates are unknown. Here, we investigate possible neural correlates of nightmare severity in a sample of individuals who frequently recall nightmares.Methods:Our principal measure of nightmare severity is nightmare distress as indexed by the Nightmare Distress Questionnaire (NDQ), and secondary measures are retrospective and prospective estimates of frequency of recalling dysphoric dreams (DD). We used high-resolution technetium 99m ethyl cysteinate dimer single photon emission computed tomography to assess regional cerebral blood flow (rCBF) while 18 individuals who were frequent nightmare recallers viewed negative and neutral pictures from the International Affective Picture System. We correlated rCBF with NDQ scores and DD recall frequency estimates.Results:Negative correlations were observed between NDQ scores and rCBF during negative picture viewing in bilateral insula and anterior cingulate, right medial frontal gyrus, bilateral superior temporal gyrus, right inferior frontal and precentral gyri, and bilateral putamen. Retrospective DD recall correlated with rCBF activity primarily in regions overlapping those related to NDQ scores. Prospective DD recall was only weakly related to rCBF. Results for the neutral condition overlapped partially with those for the negative condition; in particular, NDQ and retrospective DD recall were related to rCBF in medial prefrontal and anterior cingulate gyri.Conclusions:Results point to a possible overlap in brain mechanisms involved in nightmare dysphoria (during sleep) and distress (during wakefulness) among individuals who frequently recall nightmares. They provide partial support for a neurocognitive model of nightmares.Commentary:A commentary on this article appears in this issue on page 179.Citation:Marquis LP, Julien SH, Baril AA, Blanchette-Carrière C, Paquette T, Carr M, Soucy JP, Montplaisir J, Nielsen T. Nightmare severity is inversely related to frontal brain activity during waking state picture viewing. J Clin Sleep Med. 2019;15(2):253–264.BRIEF SUMMARYCurrent Knowledge/Study Rationale: There is growing evidence that nightmares cause clinically significant distress and may be a risk factor for psychopathology and suicidal behavior. However, there is a paucity of research on the neural mechanisms of nightmares, especially of nontraumatic nightmares. We therefore studied individuals who frequently recalled nightmares using single photon emission computed tomography.Study Impact: This study is among the first to investigate the neural correlates of disturbed dreaming, and the first to use nightmare frequency and distress severity measures. Negative correlations between nightmare severity and anterior cingulate/medial prefrontal cortices activity partially support a neurocognitive model emphasizing prefrontal regulatory mechanisms, whereas secondary results suggest that reduced activity in a wide brain network may be involved in nightmare production.INTRODUCTIONNightmares are a frequent comorbid symptom of various psychopathologies, including mood and anxiety disorders and, most notably, posttraumatic stress disorder (PTSD).1 Their prevalence in such conditions varies from, for example, 15% in anxiety disorders to 67% in PTSD. The presence of nightmares as a comorbid symptom tends to signal a greater severity of subjective distress in such pathologies. For example, in one large psychiatric sample (n = 498), patients with frequent nightmares had more severe symptoms than did patients without nightmares.2 Even in the general population, nightmare occurrence and severity are associated with increased worry, depersonalization, hallucinatory experiences, and paranoia.3Frequent nightmares observed in the absence of other clinically significant psychopathology are termed “Nightmare Disorder.” A Nightmare Disorder diagnosis can be given when nightmares cause severe distress and impair daytime functioning.4 The prevalence of the disorder in adults is estimated to be between 1% and 8%. Typically, having nightmares at least weekly is considered clinically significant.4,5It remains unknown whether trait or state factors are more critical to the severity of nightmares,6,7 and a better understanding of these factors' contributions could greatly affect treatment strategies that aim to reduced waking distress. State factors (eg, day-to-day changes in the presence of negative events that exceed emotion regulation capacity) are thought to be more closely associated with the frequency of disturbing dreams, whereas trait factors (eg, a general disposition toward high negative affect and emotional reactivity) may be more likely to explain nightmare-induced distress, that is, their intensity, effect on daytime functioning,8 and association with psychopathology.9,10 The widely used concept of nightmare distress (NMD) is one such trait factor that captures the severity of waking distress associated with nightmares and is related to psychopathology and motivation to seek treatment.11,12There is limited but consistent evidence that nightmares can affect daytime functioning. Among the rare studies that prospectively measure the effect of nightmares, Köthe and Pietrowsky13 compared self-reports of emotions following nights with and without nightmares and showed that participants felt more agitated, physically aroused, anxious and sad, less able to concentrate, less cheerful, and less self-confident among other differences on days after nightmares. Similarly, using a prospective design comparing nights with and without nightmares, Lancee and Schrijnemaekers14 found that nightmares produce daytime distress.The potential for nightmares to induce lasting distress may be critical to the finding that nightmares are a risk factor for self-harm behaviors. In one study, prospectively measured nightmares were associated, in a unidirectional fashion, to a fourfold increase in self-harmful thoughts and behaviors; the relationship was mediated by postsleep negative affect.15 Accordingly, the personality trait of NMD can be seen as a general disposition to react to nightmares with these types of dysphoric responses, that is, increased negative affect, suicidal ideation, self-harm and, possibly, maladaptive coping mechanisms. Closer study of the mechanisms and neural structures causing nightmares to negatively affect waking behavior could thus have substantial clinical utility, for example, in suggesting types of maladaptive waking behaviors to target with therapy.Our neurocognitive model7,16 proposes that nightmares arise from disturbances in a fear extinction function of normal dreaming, a function that relies on a limbic-prefrontal emotion regulation network comprising primarily medial prefrontal cortex (mPFC), anterior cingulate cortex (ACC), hippocampus, and amygdala. These regions, which are active in rapid eye movement (REM) sleep, have well-documented emotion regulation functions, whether in rodents or in humans. According to the cross-state continuity assumption of the model, emotion regulation works in a similar fashion across states and participants suffering from nightmares are therefore likely to demonstrate corresponding alterations in daytime functioning.7,16By this account, nightmares may result from the disturbance of downregulation, by mPFC and ACC, of fear processes governed by amygdala and hippocampus. In fact, limited evidence implicates ACC17 and mPFC18 in nightmare frequency, but these early reports do not directly consider nightmare-induced distress. Further, disturbances of this network could stem from adverse experiences occurring during a critical early developmental period, leading to premature development of emotion regulation abilities and remembrance of normally forgotten, distressful memories.19 Accordingly, the experience of nightmares and a concomitant breakdown of the affect regulation function of dreaming may sensitize an individual to negative affect over and above the effects of diminished sleep quality or duration. This mechanism may help explain the negative effects of nightmares on waking emotions.13–15,20In summary, closer study of the waking state neural correlates of nightmare severity could lead to a better understanding of how nightmares affect emotions and behavior during wakefulness and, thus, how nightmares influence various psychopathologies. Such work could contribute to more effective strategies for preventing nightmares, for coping with and treating nightmares, and for assessing putative neurobiological correlates of nightmares and their successful treatment. Thus, the goal of this study was to assess relationships between nightmare severity and brain activity during an induced dysphoric mood. Studying participants during waking state is coherent both with the definition of NMD as a daytime reaction to nightmares11,12 and with the cross-state continuity assumption of the neurocognitive model of nightmares.7,17,18Objectives and HypothesesOur objective was to investigate if nightmare severity—and NMD in particular—is related to brain activity during wakefulness. Our primary endpoint was NMD, as measured by the Nightmare Distress Questionnaire (NDQ12), and our secondary measures were retrospective and prospective recall of dysphoric dreams. Our secondary objective was to investigate whether dysphoric dream recall frequency measures are related to brain activity and if these relationships differ from those of nightmare distress.Given the scarcity of brain imaging studies of individuals who are frequent nightmare recallers, we used the neurocognitive model of nightmares16 to formulate hypotheses. Specifically, we predicted that nightmare distress would be correlated with reduced activity measured via regional cerebral blood flow (rCBF) in bilateral prefrontal areas known to downregu-late fear processes governed by the amygdala, that is, in the ACC and mPFC. We also expected NMD to be correlated with reduced activity in the hippocampus and increased activity in the amygdala.METHODSParticipantsWe recruited participants with frequent nightmare recall who underwent high-resolution single photon emission computed tomography (SPECT) after receiving a radiotracer injection of technetium 99m ethyl cysteinate dimer (99mTc-ECD) during the viewing of negatively and neutrally valenced pictures. Participants were part of a larger project on the neural correlates of nightmares, preliminary findings for which have been presented at conferences and/or published as abstracts.18,21Participants were recruited by advertisements on local university campuses, through our laboratory's website and by word of mouth. They were aged 18 to 35 years and were fluent in English or French. Each underwent a telephone screening interview and were included if they: (1) reported recalling at least two nightmares or bad dreams (dysphoric dreams without awakening) per week; (2) did not report presence of sleep disorders (eg, isolated sleep paralysis, night terrors, narcolepsy); (3) reported at least average sleep quality and sleeping at least 6 h/night; (4) reported consumption of fewer than 10 alcoholic beverages/wk, not using drugs except marijuana (1/mo or less) and having a daily caffeine intake equivalent to 3 cups of coffee or less; (5) did not report recent (past 6 months) traumatic experiences; (6) did not report psychiatric or medical conditions susceptible to interact with dreaming or with their ability to safely undergo SPECT scanning; and (7) took no medications other than oral contraceptives. For more details about screening, see Marquis et al.22Our initial sample included 23 individuals who were frequent nightmare recallers (3 males, 20 females). Two participants reported a traumatic event on the Posttraumatic Stress Disorder Checklist for DSM-5 and scored over the recommended cutoff point for PTSD, and two were mildly depressed (Beck Depression Inventory-II [BDI-II] score > 14); these were excluded from further analyses. Another participant had an abnormality on neuroimaging, so these SPECT images were excluded from analyses. The final sample included 18 right-handed individuals who were frequent nightmare recallers (3 males, 15 females).The study was approved by the Research Center's ethics and scientific committees. Participants provided written informed consent after being given a complete description of the study protocol. They were compensated financially for time spent in the laboratory, parking/public transit, and meal expenses.ProcedureParticipants completed questionnaires including, but not limited to, those listed in the Questionnaires section. Following their first laboratory visit, participants started home sleep-dream logs and had brain scans scheduled for 1 and 2 weeks later. When they returned to the laboratory (Figure 1), they were fitted with a forearm catheter and underwent the negative or the neutral picture viewing condition. At picture #30 the radiotracer was injected. Picture viewing was followed by a short, humorous video to stabilize mood, then by the SPECT scan, and then participants could leave. They returned 1 week later for the second scan (same procedure: see Figure 1) preceded by the other picture viewing condition. Half of the participants completed the neutral condition first, the other half the negative condition.Figure 1: Protocol for picture viewing conditions and SPECT scan.SPECT = single photon emission computed tomography.Download FigureQuestionnairesParticipants completed the State-Trait Anxiety Inventory (STAI);23 BDI-II;24 NDQ;12 and Posttraumatic Stress Disorder Checklist for DSM-5.25 The BDI-II cutoff point of 14 to 19 for slight depression26 was applied in screening participants.Nightmare Severity MeasuresNightmare DistressThe NDQ is a 13-item questionnaire (response scale of 1 to 5, from never (1) to always (5); total score of 13 to 65) assessing various forms of waking distress associated with nightmares. Initial validation studies11,12 have found adequate internal consistency, with Cronbach alpha coefficients of .80–.90. Belicki11,12 found both nightmare frequency and distress to be associated with interest in therapy for nightmares, with higher correlations for nightmare distress. Although nightmare distress is related to nightmare frequency,11,27 it is primarily associated with psychopathology.10,11Home Sleep-Dream LogParticipants kept daily logs for 2 consecutive weeks beginning on the morning following the first laboratory visit. They used an interactive voicemail system28 for recording dream reports and rating sleep features (quality, number of hours, napping, number of awakenings) and dream content (recall clarity, positive and negative emotion, whether dream awakened them); most ratings used Likert scales (1 to 9). Prospective dream recall (0/1) was scored as successful when recall clarity was ≥ 1/9; bad dream recall (0/1) when negative emotion was ≥ 5/9; a nightmare (0/1) when negative emotion was ≥ 5/9 and the dream caused an awakening. The dysphoric dream recall measure was the sum of the bad dream and nightmare recall measures. Results were computed to obtain weekly prospective frequencies for dreams, bad dreams, nightmares, and dysphoric dreams (Table 1).Table 1 Participant characteristics.Table 1 Participant characteristics.Retrospective MeasuresRetrospective measures were derived from the initial telephone screening (conducted up to several weeks before the laboratory visit) and computed as weekly frequencies of recalling dreams, bad dreams, nightmares, and dysphoric dreams (Table 1).Three measures were selected for assessment in relation to brain activity: NDQ total score and weekly dysphoric dream recall measured both prospectively (prosDD) and retrospectively (retroDD).Experimental Condition: International Affective Picture SystemDuring radiotracer injection for the SPECT scan (see next paragraphs), participants viewed negatively or neutrally valenced International Affective Picture System (IAPS) pictures.29 Each participant viewed sets of both negative and neutral IAPS pictures (in counterbalanced order) in two separate brain imaging sessions scheduled 1 week apart. IAPS is a stimulus set that reliably elicits mood changes30 with each picture having been rated normatively for its emotional valence (negative versus positive) and emotional arousal (intensity). Common themes for the selected negative pictures corresponded in a general way to themes typically reported to occur in nightmares, for example, actual or threatened violence between humans, dangerous animals, and wounded or dead animals and humans. Common themes for the selected neutral pictures included friendly interactions between humans, nonthreatening animals, and wilderness landscapes.Selected negative pictures had a normatively scored mean of 2.66 (standard deviation [SD] = 0.72) on the 9-point valence scale (1 = negative, 9 = positive) and a mean of 5.67 (SD = 0.74) on the 9-point arousal scale (1 = calm, 9 = excited). Selected neutral pictures had a normatively scored mean valence of 6.88 (SD = 0.9) and a mean arousal of 4.0 (SD = 0.79). Pictures were ordered so that mean valence and arousal before and after picture #30 (timing of radio-tracer injection) were similar. Using Inquisit software (version 4, Millisecond Software), participants were first shown 10 practice pictures with normative mean valence of 6.92 (SD = 1.11) and mean arousal of 4.57 (SD = 0.60) and then 48 negative or neutral pictures for 10 seconds each, with a 1-second inter-picture interval. They were shown a humor-istic 3-minute video (Simon's Cat, YouTube) to normalize mood after radiotracer uptake; because most radiotracer uptake occurs in the 2 minutes post-injection,31 this did not influence neuroimaging results.To evaluate stimulus efficacy, participants rated their emotional valence and arousal on scales of 1 to 9 using the Self-Assessment Manikin32 and rated 11 emotions on a modified Differential Emotions Scale33 on a scale of 1 to 5 (1 = very little, 5 = very strongly) four times: (1) before the practice, (2) after the practice, (3) after viewing 48 negative pictures, and (4) after viewing the humoristic video (Figure 1).Brain ImagingTechnetium 99m Ethyl Cysteinate Dimer (99mTc-ECD) SPECT Image AcquisitionSPECT image acquisition and analysis parameters were similar to those used by Baril el al.,34 and used the same high-resolution (2.5 mm full-width half-maximum) NeuroFOCUS scanner (NeuroPhysics, Shirley, Massachusetts, United States) although with a different radiotracer (99mTc-ECD) compared to 99mTc-HMPAO in Baril et al. At IAPS picture #30, participants were given a dose of 750 megabecquerel of 99mTc-ECD followed by a 30-cc saline flush. Ten minutes postinjection, participants underwent a standard 30-minute image acquisition sequence. The cerebellum was excluded from analysis. Acquisitions all occurred between 9:30 am and 4:45 pm, according to participant preferences.SPECT Image AnalysisImages were inspected visually for quality. We used SPM8 (Statistical Parametric Mapping 8, Wellcome Trust Centre for Neuroimaging, Institute of Neurology, University College London, United Kingdom) with MatLab (version 8.6, The Mathworks, Natick, Massachusetts, United States) to preprocess images (coregistration and normalization to SPECT template, smoothing of 14 mm full-width half-maximum). rCBF values from each image were normalized for their individual global mean signal. Final voxel size was 2 × 2 × 2 mm.Statistical AnalysesDemographics, Questionnaires, Screening Interview, Home Sleep-Dream Log, Picture Viewing RatingsDistributions of these measures were examined for normality and descriptive statistics generated with SPSS 20 (IBM Inc., Armonk, New York, United States).SPECT Correlational AnalysesUsing a multiple regression design in SPM8 we separately correlated rCBF values sampled during viewing of negative and neutral pictures with NDQ scores, retroDD, and prosDD using a statistical threshold of P < .005 (uncorrected) and a cluster extent threshold of k > 100, which corresponds to the minimal number of contiguous voxels necessary for the cluster to be considered significant. Analyses were performed on every voxel of gray matter using a mask. The combination of liberal P values and high cluster-extent threshold is optimal for localizing seizure-onset zones in patients with epilepsy.35PickAtlas software (version 3.0.5)36 was used to identify significant regions from the ICBM atlas37 and to create a gray matter mask. Significant clusters were displayed on the magnetic resonance imaging (MRI) template included in the MRI-cron program ( http://people.cas.sc.edu/rorden/mricron/index.html). MRIcron was also used to generate figures.RESULTSDemographics, Questionnaires, Screening Interview, Sleep-Dream Log, Picture Viewing RatingsMeans and SDs for age, IAPS picture viewing ratings, questionnaire scores, and all retrospective and prospective dream recall frequencies are reported in Table 1. Pearson correlations revealed that NDQ scores were not associated with either STAI-T or BDI-II scores (P > .20), and STAI-T and BDI-II scores were not intercorrelated (P > .20). Spearman correlations were computed between questionnaire variables and dream frequency measures, as the dream frequency measures had skewed distributions (Table 2).Table 2 Spearman correlations for questionnaire responses and retrospective and prospective dream recall frequencies.Table 2 Spearman correlations for questionnaire responses and retrospective and prospective dream recall frequencies.SPECT Correlational Analyses: Negative PicturesThere was a preponderance of negative correlations between rCBF and NDQ, retroDD and prosDD. First, there were negative correlations between rCBF and NDQ in several brain regions (Table 3 and Figure 2), including bilateral cingulate gyrus, right medial frontal gyrus, bilateral superior temporal gyrus, right inferior frontal gyrus, left precentral gyrus, and bilateral anterior insula and putamen, but no positive correlations.Table 3 Localization of hypoperfused regions associated with nightmare severity measures during negative picture viewing.Table 3 Localization of hypoperfused regions associated with nightmare severity measures during negative picture viewing.Figure 2: Coronal and axial multislice view of hypoperfused regions associated with nightmare distress during negative picture viewing.Color code: cyan = right medial frontal gyrus and left cingulate gyrus, green = right superior temporal gyrus, red = right putamen, fuschia = left superior temporal gyrus, yellow = right inferior frontal gyrus, white with red border = left putamen and insula, white with gray border = right insula, white with green border = left precentral gyrus and insula, white with blue border = right anterior cingulate gyrus and medial frontal gyrus. Significant regions were obtained with the following combination of statistical thresholds: peaks at P < .005 within clusters > 100.Download FigureSimilarly, there were negative correlations between rCBF and retroDD in left anterior cingulate gyrus, bilateral medial frontal gyrus, left middle frontal gyrus, bilateral inferior frontal gyrus, and left putamen and insula (Table 3 and Figure 3), but no positive correlations.Figure 3: Coronal and axial multislice view of hypoperfused regions associated with retrospective dysphoric dream recall frequency during negative picture viewing.Color code: cyan = middle frontal gyrus, green = left anterior cingulate and right frontal medial prefrontal gyri, red = left medial prefrontal gyrus, fuchsia = left middle inferior frontal gyrus, yellow = right inferior frontal gyrus, white with red border = left inferior temporal gyrus, white with gray border = left inferior frontal gyrus, putamen and insula. Significant regions were obtained with the following combination of statistical thresholds: peaks at P < .005 within clusters > 100.Download FigureFinally, there were negative correlations between rCBF and prosDD in left middle frontal gyrus and left lateral orbital frontal gyrus (Table 3) and only minimal positive correlations in right lingual gyrus (k = 309, P < .001; X = 12, Y = −54, Z = 2, Broadmann area (BA) 18, t = 5.58) and right middle temporal gyrus (k = 165, P < .001; X = 58, Y = 8, Z = −20, BA 21, t = 3.82).SPECT Correlational Analyses: Neutral PicturesAs for the negative pictures, there was a preponderance of negative correlations between rCBF and NDQ, retroDD and prosDD. First, there were negative correlations between rCBF and NDQ in several brain regions (Table 4 and Figure 4), including bilateral anterior cingulate gyrus, left medial frontal gyrus, bilateral superior temporal gyrus, left middle temporal gyrus, postcentral gyrus, thalamus and putamen, and a positive correlation in left middle occipital gyrus (k = 162, P < .001; X = −24, Y = −94, Z = 6, BA 18, t = 3.93).Table 4 Localization of hypoperfused regions associated with nightmare severity measures during neutral picture viewing.Table 4 Localization of hypoperfused regions associated with nightmare severity measures during neutral picture viewing.Figure 4: Coronal and axial multislice view of hypoperfused regions associated with nightmare distress during neutral picture viewing.Color code: cyan = right superior temporal gyrus, green = left putamen, red = bilateral anterior cingulate gyrus, fuschia = left superior and middle temporal gyri, yellow = left superior temporal and postcentral gyri, white with red border = right thalamus, white with gray border = left anterior cingulate gyrus and medial frontal gyrus. Significant regions were obtained with the following combination of statistical thresholds: peaks at P < .005 within clusters > 100.Download FigureAdditionally, there were negative correlations between rCBF and retroDD in right anterior cingulate gyrus, right me-dial frontal gyrus, left insula, and inferior, middle, and superior frontal gyri (Table 4). No positive correlations were observed.Finally, there was a negative correlation between rCBF and prosDD in left middle frontal gyrus (Table 4) and positive correlations in left anterior cingulate gyrus (k = 121, P < .001; X = −2, Y = 22, Z = −6, BA 24, t = 4.07), right posterior cingulate gyrus (k = 191; P < .001, X = 10, Y = −46, Z = −6, BA 29, t = 3.85; P < .005, X = 12, Y = −62, Z = 12, BA 30, t = 3.43 and P < .001, X = 6, Y = −70, Z = 8, BA 30, t = 3.01) and right middle temporal gyrus (k = 106, P < .005; X = 50, Y = 6, Z = −18, BA 21, t = 3.60).DISCUSSIONMain SPECT FindingsWe aimed to evaluate whether nightmare severity, that is, NDQ and retrospective and pros" @default.
- W2912551200 created "2019-02-21" @default.
- W2912551200 creator A5000732938 @default.
- W2912551200 creator A5009418118 @default.
- W2912551200 creator A5031039978 @default.
- W2912551200 creator A5036473548 @default.
- W2912551200 creator A5047169531 @default.
- W2912551200 creator A5056249680 @default.
- W2912551200 creator A5062988766 @default.
- W2912551200 creator A5063275477 @default.
- W2912551200 creator A5086713417 @default.
- W2912551200 date "2019-02-15" @default.
- W2912551200 modified "2023-10-01" @default.
- W2912551200 title "Nightmare Severity Is Inversely Related to Frontal Brain Activity During Waking State Picture Viewing" @default.
- W2912551200 cites W182973379 @default.
- W2912551200 cites W1897424475 @default.
- W2912551200 cites W1965623943 @default.
- W2912551200 cites W1968715116 @default.
- W2912551200 cites W1973230171 @default.
- W2912551200 cites W1974743833 @default.
- W2912551200 cites W1987574933 @default.
- W2912551200 cites W1990137058 @default.
- W2912551200 cites W1991742699 @default.
- W2912551200 cites W1997603362 @default.
- W2912551200 cites W1997801013 @default.
- W2912551200 cites W2021919067 @default.
- W2912551200 cites W2028897182 @default.
- W2912551200 cites W2035146528 @default.
- W2912551200 cites W2041879673 @default.
- W2912551200 cites W2050618347 @default.
- W2912551200 cites W2063221126 @default.
- W2912551200 cites W2074840506 @default.
- W2912551200 cites W2077083873 @default.
- W2912551200 cites W2090459429 @default.
- W2912551200 cites W2094500507 @default.
- W2912551200 cites W2110747672 @default.
- W2912551200 cites W2111464173 @default.
- W2912551200 cites W2141747781 @default.
- W2912551200 cites W2142322501 @default.
- W2912551200 cites W2143895814 @default.
- W2912551200 cites W2167557160 @default.
- W2912551200 cites W2339259158 @default.
- W2912551200 cites W2411089591 @default.
- W2912551200 cites W2527576709 @default.
- W2912551200 cites W2545448426 @default.
- W2912551200 cites W2620166550 @default.
- W2912551200 cites W2664764853 @default.
- W2912551200 cites W2732740969 @default.
- W2912551200 cites W2735872510 @default.
- W2912551200 cites W2766757463 @default.
- W2912551200 cites W2770497909 @default.
- W2912551200 doi "https://doi.org/10.5664/jcsm.7628" @default.
- W2912551200 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/6374093" @default.
- W2912551200 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/30736883" @default.
- W2912551200 hasPublicationYear "2019" @default.
- W2912551200 type Work @default.
- W2912551200 sameAs 2912551200 @default.
- W2912551200 citedByCount "12" @default.
- W2912551200 countsByYear W29125512002019 @default.
- W2912551200 countsByYear W29125512002020 @default.
- W2912551200 countsByYear W29125512002021 @default.
- W2912551200 countsByYear W29125512002022 @default.
- W2912551200 countsByYear W29125512002023 @default.
- W2912551200 crossrefType "journal-article" @default.
- W2912551200 hasAuthorship W2912551200A5000732938 @default.
- W2912551200 hasAuthorship W2912551200A5009418118 @default.
- W2912551200 hasAuthorship W2912551200A5031039978 @default.
- W2912551200 hasAuthorship W2912551200A5036473548 @default.
- W2912551200 hasAuthorship W2912551200A5047169531 @default.
- W2912551200 hasAuthorship W2912551200A5056249680 @default.
- W2912551200 hasAuthorship W2912551200A5062988766 @default.
- W2912551200 hasAuthorship W2912551200A5063275477 @default.
- W2912551200 hasAuthorship W2912551200A5086713417 @default.
- W2912551200 hasBestOaLocation W29125512001 @default.
- W2912551200 hasConcept C118552586 @default.
- W2912551200 hasConcept C15744967 @default.
- W2912551200 hasConcept C169760540 @default.
- W2912551200 hasConcept C2777992674 @default.
- W2912551200 hasConcept C2779320081 @default.
- W2912551200 hasConcept C522805319 @default.
- W2912551200 hasConcept C548259974 @default.
- W2912551200 hasConcept C71924100 @default.
- W2912551200 hasConceptScore W2912551200C118552586 @default.
- W2912551200 hasConceptScore W2912551200C15744967 @default.
- W2912551200 hasConceptScore W2912551200C169760540 @default.
- W2912551200 hasConceptScore W2912551200C2777992674 @default.
- W2912551200 hasConceptScore W2912551200C2779320081 @default.
- W2912551200 hasConceptScore W2912551200C522805319 @default.
- W2912551200 hasConceptScore W2912551200C548259974 @default.
- W2912551200 hasConceptScore W2912551200C71924100 @default.
- W2912551200 hasIssue "02" @default.
- W2912551200 hasLocation W29125512001 @default.
- W2912551200 hasLocation W29125512002 @default.
- W2912551200 hasLocation W29125512003 @default.
- W2912551200 hasLocation W29125512004 @default.
- W2912551200 hasLocation W29125512005 @default.
- W2912551200 hasOpenAccess W2912551200 @default.
- W2912551200 hasPrimaryLocation W29125512001 @default.