Post by Admin on Nov 26, 2020 17:20:03 GMT
The Functional Role of Dreaming in Emotional Processes
www.frontiersin.org/articles/10.3389/fpsyg.2019.00459/full
Dream experience (DE) represents a fascinating condition linked to emotional processes and the human inner world. Although the overlap between REM sleep and dreaming has been overcome, several studies point out that emotional and perceptually vivid contents are more frequent when reported upon awakenings from this sleep stage. Actually, it is well-known that REM sleep plays a pivotal role in the processing of salient and emotional waking-life experiences, strongly contributing to the emotional memory consolidation. In this vein, we highlighted that, to some extent, neuroimaging studies showed that the processes that regulate dreaming and emotional salience in sleep mentation share similar neural substrates of those controlling emotions during wakefulness. Furthermore, the research on EEG correlates of the presence/absence of DE and the results on EEG pattern related to the incorporated memories converged to assign a crucial role of REM theta oscillations in emotional re-processing. In particular, the theta activity is involved in memory processes during REM sleep as well as during the waking state, in line with the continuity hypothesis. Also, the gamma activity seems to be related to emotional processes and dream recall as well as to lucid dreams. Interestingly, similar EEG correlates of DE have been found in clinical samples when nightmares or dreams occur. Research on clinical samples revealed that promoting the rehearsal of frightening contents aimed to change them is a promising method to treat nightmares, and that lucid dreams are associated with an attenuation of nightmares. In this view, DE can defuse emotional traumatic memories when the emotional regulation and the fear extinction mechanisms are compromised by traumatic and frightening events. Finally, dreams could represent a sort of simulation of reality, providing the possibility to create a new scenario with emotional mastery elements to cope with dysphoric items included in nightmares. In addition, it could be hypothesized that the insertion of bizarre items besides traumatic memories might be functional to “impoverish” the negative charge of the experiences.
Introduction
Dreaming represents a fascinating experience linked to emotional processes so much so to be considered as a key to “access” in the human inner world (e.g., Freud's view; Freud, 1955). In the last decades, dream experience (DE) has been studied under the scientific perspective, defining dream recall as the retrieval of mental sleep elaboration with a different level of complexity and fragmentation, reported after awakenings (Fagioli, 2002).
For years, researchers considered Rapid Eye Movements (REM) sleep the privileged stage to dream (Desseilles et al., 2011) as long as different criteria to collect the reports were introduced (Foulkes, 1962). Actually, DE could be categorized as dream-like or thought-like on the basis of dream contents (Cavallero et al., 1992), including in the first the mentation related to higher emotionality and narrative contents and in the second, to more fragmented elements (Cipolli et al., 2017). Several studies confirmed that dreaming can occur both during REM and NREM sleep (Foulkes, 1962; Rechtschaffen et al., 1963; Monroe et al., 1965; Pivik and Foulkes, 1968; Taub, 1971; Cavallero et al., 1992; Antrobus et al., 1995; Stickgold et al., 2001). In particular, the DEs were obtained in up to 50% of cases upon awakening from NREM sleep, especially from stage 2 (Foulkes, 1962; Pivik and Foulkes, 1968; Nielsen, 2000) and REM sleep suppression pharmacologically-induced did not affect dream recall (Landolt et al., 2001; Oudiette et al., 2012).
Although the exclusive relation between REM sleep and dream has been overcome (Scarpelli et al., 2015a), several studies point out that emotional and perceptually vivid contents are more frequent when reported upon awakenings from REM sleep (Foulkes et al., 1988; Nielsen et al., 1991; Merritt et al., 1994; Hobson et al., 2000; Kahn et al., 2002). More in general, REM sleep plays a pivotal role in the processing of emotional events and several studies showed that the consolidation of emotional memories occurs in this sleep stage (e.g., Lara-Carrasco et al., 2009; Nishida et al., 2009). In addition, experimental deprivation of REM sleep has been demonstrated to compromise the consolidation of emotional stimuli (Cartwright et al., 1975; Wagner et al., 2001; Lara-Carrasco et al., 2009; Spoormaker et al., 2014).
Dream contents including negative emotions (e.g., anxiety and fear) are more frequent than positive ones (Foulkes et al., 1988; Nielsen et al., 1991; Merritt et al., 1994; Fosse et al., 2001), and they are often related to waking-life experiences (Stickgold et al., 2001; Wamsley et al., 2010; Eichenlaub et al., 2017; Vallat et al., 2017). In this regard, the existing empirical evidence highlighted that the neural activation of emotional-limbic (Nir and Tononi, 2010) and reward systems (Perogamvros and Schwartz, 2012) during REM-DE contributes to the offline reprocessing of emotions and associative learning (Perogamvros et al., 2013).
Interestingly, clinical studies have also provided evidence on the potential interaction among DE, sleep alterations and affective disorders (Levin and Nielsen, 2007; Nielsen and Levin, 2007; Schredl, 2011). For instance, mood disturbances frequently appear along with sleep modifications involving REM sleep (Benca et al., 1992, 1997; Walker and van der Helm, 2009). More directly, abundant dreams or nightmares are related to REM sleep abnormalities and psychiatric disorders (Cartwright et al., 2003; Modell et al., 2005; Agargun et al., 2007; Schredl et al., 2009; Sjöström et al., 2009; Marinova et al., 2014; Nakajima et al., 2014).
Taking into account the impossibility to directly access to the DE, most studies in the last decades were focused on the retrieval of dream contents upon awakenings (Scarpelli et al., 2015a). Specifically, several investigations using spontaneous or provoked awakenings in the laboratory aimed to find the electrophysiological (EEG) patterns related to dream recall, considering it as a form of episodic-declarative memory (Mangiaruga et al., 2018).
In keeping with some earlier reviews (e.g., Revonsuo, 2000; Revonsuo et al., 2015), it could be hypothesized that DE may play a pivotal role in emotional encoding and regulation, nevertheless the specific function and the neural bases of REM-DEs were directly investigated by very few studies (De Gennaro et al., 2011; Eichenlaub et al., 2018; Vallat et al., 2018a).
In light of the above, here we reviewed the main findings about neural bases of dreaming and the link between EEG correlates of DE and emotional processing, underlying that some evidence on sleep disorders characterized by a peculiar DE, especially occurring in REM sleep, may provide a better understanding on the functional role of dreaming and its potential applications in the clinical field.
Methodological Note
Studies were identified via PUBMED queries. Key search terms included:
- “Dreaming” and “REM sleep” and “Neuroimaging” in title/abstract (62 articles)
- “Dreaming” and “REM sleep” and “Emotional processing” (22 articles);
- “Dreaming” and “EEG” in title/abstract (686 articles);
- “Dreaming” and “PTSD” in title/abstract (450 articles);
- “Dreaming” and “Emotional Memories” in title/abstract (61 articles);
- “Dreaming” and “Narcolepsy” in title/abstract (128 articles);
We grouped the identified articles in the following categories:
1. Cognitive features, emotional aspects and EEG patterns of mental activity during REM sleep;
2. Dreaming and emotional process: (a) role of REM sleep in processing salient emotional waking life experiences; (b) EEG patterns implicated in REM sleep and emotional dreams; (c) associations among traumatic events and nightmares; (d) role of lucid dreams;
3. Dreaming in narcolepsy.
We focused on the role of emotions in REM sleep, excluding non-English articles. All the articles resulting from using this method and related to our focus were included. Following these criteria, we identified 88 publications which were estimated to be of interest for further examination.
Neural Bases of Dreaming and Emotional Processing
A large body of evidence showed that the regions implicated in emotional processes during wakefulness are also responsible for the neurophysiological background of REM sleep that can explain some qualitative features of DE (Maquet et al., 1996; Nir and Tononi, 2010; De Gennaro et al., 2011, 2016; Eichenlaub et al., 2014; Vallat et al., 2018a). In this section, we highlighted the commonalities between the neural bases of REM-DE and emotional processing.
Earlier studies on anatomical correlates of dream alterations by brain lesions studies confirmed the central role of specific cortical and subcortical areas in DE. The main findings point out that two specific cortical systems underlie dreaming. Firstly, a posterior system involves the Temporo-Parieto-Occipital Junction (TPJ) and lesions located in this region alter both waking mental imagery and sleep (Solms, 1997, 2000). The anterior system includes the ventromedial prefrontal cortex (vmPFC) and the white matter surrounding the anterior horns of the lateral ventricles and its damage seems related to dream loss (Solms, 1997, 2000, 2011). It should be noted that many of the afferent and efferent fibers of these two systems are connected with the limbic system and the ventromesial frontal white matter seems to take a part in the interplay between basal forebrain and limbic structures (De Gennaro et al., 2012, 2016).
More recently, human neuroimaging studies have gained a crucial role in dream research. Positron Emission Tomography (PET) and functional Magnetic Resonance Imaging (fMRI) were used especially in the measurement of functional changes in the brain during REM sleep. It has been demonstrated that the distribution of brain activity in REM sleep is not homogenous (Maquet, 2000; Nir and Tononi, 2010). Brain imaging studies found increased regional brain activity in the limbic and paralimbic structures, pontine tegmentum, thalamus and basal forebrain during REM sleep, as compared to wakefulness (Braun et al., 1997, 1998; Nofzinger et al., 1997; Maquet, 2000) and NREM sleep (Braun et al., 1997, 1998; Maquet et al., 2005). In addition, one of the first neuroimaging studies collecting dream report (Maquet et al., 1996) found a bilateral activation of the amygdala in subjects reporting DE upon awakenings from REM sleep, not providing any comparisons with non-recall conditions or awakenings from NREM sleep.
Moreover, remarkable activation of motor and premotor regions has been found in healthy subjects during REM sleep (Braun et al., 1997; Maquet et al., 2000; De Carli et al., 2016), albeit this sleep stage is characterized by muscular atonia thanks to the inhibition of spinal motor neurons by the ponto-bulbar reticular formation (Lai and Siegel, 1999). This seems in line with the observation that individuals affected by REM sleep behavior disorder (RBD) exhibit motor behaviors linked to their DE (Oudiette et al., 2009). Other brain areas are hypoactive during REM sleep compared to the waking state, such as the dorso-lateral PFC (dlPFC), precuneus, orbitofrontal cortex (OFC) and posterior cingulate gyrus (Maquet et al., 1996; Braun et al., 1997; Nofzinger et al., 1997). This evidence could explain the altered executive functions, time perception, and the lack of insight during DE (Desseilles et al., 2011).
Although it is well-known that dreaming is not restricted to REM sleep, some REM sleep features make it a privileged background for DE, especially characterized by high vividness, bizarreness and emotional load (Carr and Solomonova, 2018). In this respect, it should be underlined that most of the regions involved in emotional memory encoding and consolidation (Phelps and LeDoux, 2005; Armony, 2013) are highly activated during REM sleep. In particular, the higher activation of amygdaloid complexes, hippocampal formation and anterior cingulate cortex (ACC) in REM sleep than in wake and/or NREM sleep may explain the emotional intensity of DE reported upon REM sleep awakenings (Maquet et al., 1996; Braun et al., 1997; Nofzinger et al., 1997; Desseilles et al., 2011; Corsi-Cabrera et al., 2016). Some authors suggested that these structures may be responsible for the reprocessing and consolidation of emotional experiences during REM sleep (Hobson and Pace-Schott, 2002; van der Helm et al., 2011; Deliens et al., 2014).
According to the idea that DE play a role in the emotional processing, more recently awake fMRI measures revealed that subjects experiencing frequent fear during DE have a higher activation of mPFC and reduced activation in the insula, amygdala and midcingulate cortex when faced with aversive stimuli (Sterpenich et al., 2019). The results are consistent with the view that mPFC should exert an inhibitory control on fear expression by reducing amygdala activity (Phelps et al., 2004). A large amount of dream reports were collected at home and the presence of fear in DE has been considered as a trait-like factor of the participants (Sterpenich et al., 2019).
These results are consistent with the continuity hypothesis, namely the possibility that dreams and wakefulness shared similar mechanisms (Schredl, 2003, 2009; Scarpelli et al., 2015a,b; Mangiaruga et al., 2018; Sterpenich et al., 2019).
In fact, neuroimaging data in waking state have provided strong support for a crucial role of the aforementioned regions in emotional processing (Phan et al., 2004). It has been proposed that in the waking brain subcortical regions such as amygdala, nucleus accumbens, locus coeruleus, pulvinar, along with the ACC and OFC are involved in conscious and unconscious emotional processing (Morris et al., 1998; Vuilleumier et al., 2001; Kirov et al., 2012). Specifically, the unconscious perception of emotional stimuli has been related with the functional integrity of subcortical areas that receive an executive cortical feedback from the cortical network (e.g., dlPFC, OFC, and posterior cingulate cortex) only during wakefulness (Tamietto and De Gelder, 2010).
It is well-known that amygdala plays a pivotal role in emotional regulation. Several studies found that this structure is responsible for detecting, generating, and maintaining fear-related emotions (for a review see Phan et al., 2004). Moreover, the amygdala is important in the coordination of adequate responses to threat and danger and it has also been demonstrated that is critical also for response to stimuli that predict positive and negative future outcomes (Paton et al., 2006), allowing the organism to learn more about a stimulus and organize adaptive behavior (Maren, 2011). Specifically, the amygdala, thanks to the connections with the hippocampus, thalamus, mPFC, and ACC, is involved in control of the encoding and retrieval of affective memories and the physical expression of emotions during the waking state (Misane et al., 2005). Also, the hippocampus has an established role in emotional memory encoding and retrieval. Along with the amygdala, the hippocampal formation mediates the processing and execution of fear memories (Phelps, 2004).
Structural brain imaging approaches offer the possibility to focus on ultrastructural, anatomical measures and inter-individual variability associated with DE. This perspective allows researchers to overcome one of the intrinsic limitations of the study of dreams, concerning the difficult to exactly define the time-coupling between the sleep stages and the occurrence of DE. Neuroanatomical parameters provide a relatively stable measure to account for some trait-like features of DE. In this vein, the correlations between inter-individual differences in quantitative/qualitative features of DE reported by subjects and structural parameters of limbic areas were observed (De Gennaro et al., 2011, 2016). Once again, the regions engaged in emotional encoding and consolidation showed a strong relationship with dream contents, consistently with the studies revealing that the structural characteristics of hippocampus and amygdala are associated with cognitive and emotional processing in waking tasks (Maguire et al., 2000; Bohbot et al., 2007; Iaria et al., 2008). Specifically, microstructural measures were related to the emotional load, bizarreness and vividness of DE reported. For what concerns emotional intensity, a relation was found between lower structural integrity of the left amygdala and decreased emotional load. Moreover, emotionality was linked to a smaller volume of the left hippocampus and larger volume of the right hippocampus. Also, dream contents characterized by high bizarreness were associated with smaller left amygdala, smaller right hippocampus and lower mean diffusivity of the right amygdala.
These observations were partially replicated on patients with Parkinson's Disease (PD), pointing that PD patients did not differ from healthy subjects with respect to sleep and dream characteristics or the neuroanatomical measures and confirming that vividness is related to the amygdalae and also the thickness of the left mPFC. It should be noted that, along with amygdala, mPFC plays a key role in the acquisition, consolidation and retrieval of fear memory and, specifically, modulates fear learning and extinction (Marek et al., 2013).
Moreover, this study on PD focused on the role of the dopaminergic system in DE. It has been found that higher dopamine agonist dosage was associated with qualitatively impoverished DE, as indicated by reduced bizarreness and emotional load levels (De Gennaro et al., 2016). It is worth noting that one of the main regulator of mesolimbic-dopamine network is the mPFC that makes direct and indirect connections to the amygdala and hippocampus (Patton et al., 2013). In other words, the recent findings on PD confirmed the evidence of an essential interplay between the mPFC and mesolimbic-dopamine system for dreaming, as previously showed by data from patients who underwent prefrontal leukotomy that stopped dreaming (Solms, 2011). In fact, in these patients, the ventromedial white matter containing dopaminergic projections to the frontal lobe were disconnected (Bradley et al., 1958). The importance of mPFC in dream processes has been confirmed in another study using PET (Eichenlaub et al., 2014). They revealed that subjects who recall more frequently their DE (i.e., High Recallers—HR) showed higher regional cerebral blood flow (rCBF) in the mPFC during REM sleep and wakefulness than “Low Recallers” (LR) along with higher rCBF in the TPJ during REM, NREM sleep and wakefulness (Eichenlaub et al., 2014). The structural data obtained by MRI from the same sample of Eichenlaub et al. (2014) and from another research (Vallat et al., 2018b) were analyzed, confirming the relationship between brain anatomical structures and dream recall rate (Vallat et al., 2018a). Specifically, the anatomical measures of the mPFC were significantly different between HR and LR (i.e., increased mPFC white-matter density in HR compared to LR was found). It should be underlined that mPFC is engaged in mind representations and evaluation (for a review, see Legrand and Ruby, 2009), that could have a crucial role in the emotional processing. More in general, it has been hypothesized that mPFC is related to cognitive aspects of emotional processing, such as appraisal/identification of emotion, attention to emotion, awareness and introspection (Phan et al., 2004). Furthermore, it has been demonstrated the involvement of this region in the social emotions processing (Ruby and Decety, 2004). Recently, it has been demonstrated that traumatic experiences and pathological memories are linked to abnormal interactions between hippocampus and mPFC that are responsible to reduced mnemonic ability and decreased emotional control (Maren et al., 2013; Jin and Maren, 2015).
Moreover, it is worth noting that lesion, imaging and stimulation data revealed that the TPJ is related to emotional processing and especially to the “theory of mind,” empathy and social cognition in the waking state (Saxe and Kanwisher, 2003; Young et al., 2010; Santiesteban et al., 2012; Van Overwalle and Vandekerckhove, 2013; Jeurissen et al., 2014). In particular, TPJ contributes to the mentalizing and emotional state inference/attribution (Donaldson et al., 2015; Ye et al., 2015; Biervoye et al., 2016).
Taken together, these findings revealed that, to some extent, the processes that regulate dreaming and emotional salience in DE share similar neural substrates of those controlling emotions during wakefulness. Moreover, in the light of the results by Eichenlaub et al. (2014), one might speculate that the high dream recaller may be particularly interested in their inner world (Ioannides et al., 2009), given that salience seems to have a pivotal role in dream recall, as evaluated in older individuals measuring episodic memory recall (for a review, see Mangiaruga et al., 2018).
(rest in Link)
www.frontiersin.org/articles/10.3389/fpsyg.2019.00459/full
Dream experience (DE) represents a fascinating condition linked to emotional processes and the human inner world. Although the overlap between REM sleep and dreaming has been overcome, several studies point out that emotional and perceptually vivid contents are more frequent when reported upon awakenings from this sleep stage. Actually, it is well-known that REM sleep plays a pivotal role in the processing of salient and emotional waking-life experiences, strongly contributing to the emotional memory consolidation. In this vein, we highlighted that, to some extent, neuroimaging studies showed that the processes that regulate dreaming and emotional salience in sleep mentation share similar neural substrates of those controlling emotions during wakefulness. Furthermore, the research on EEG correlates of the presence/absence of DE and the results on EEG pattern related to the incorporated memories converged to assign a crucial role of REM theta oscillations in emotional re-processing. In particular, the theta activity is involved in memory processes during REM sleep as well as during the waking state, in line with the continuity hypothesis. Also, the gamma activity seems to be related to emotional processes and dream recall as well as to lucid dreams. Interestingly, similar EEG correlates of DE have been found in clinical samples when nightmares or dreams occur. Research on clinical samples revealed that promoting the rehearsal of frightening contents aimed to change them is a promising method to treat nightmares, and that lucid dreams are associated with an attenuation of nightmares. In this view, DE can defuse emotional traumatic memories when the emotional regulation and the fear extinction mechanisms are compromised by traumatic and frightening events. Finally, dreams could represent a sort of simulation of reality, providing the possibility to create a new scenario with emotional mastery elements to cope with dysphoric items included in nightmares. In addition, it could be hypothesized that the insertion of bizarre items besides traumatic memories might be functional to “impoverish” the negative charge of the experiences.
Introduction
Dreaming represents a fascinating experience linked to emotional processes so much so to be considered as a key to “access” in the human inner world (e.g., Freud's view; Freud, 1955). In the last decades, dream experience (DE) has been studied under the scientific perspective, defining dream recall as the retrieval of mental sleep elaboration with a different level of complexity and fragmentation, reported after awakenings (Fagioli, 2002).
For years, researchers considered Rapid Eye Movements (REM) sleep the privileged stage to dream (Desseilles et al., 2011) as long as different criteria to collect the reports were introduced (Foulkes, 1962). Actually, DE could be categorized as dream-like or thought-like on the basis of dream contents (Cavallero et al., 1992), including in the first the mentation related to higher emotionality and narrative contents and in the second, to more fragmented elements (Cipolli et al., 2017). Several studies confirmed that dreaming can occur both during REM and NREM sleep (Foulkes, 1962; Rechtschaffen et al., 1963; Monroe et al., 1965; Pivik and Foulkes, 1968; Taub, 1971; Cavallero et al., 1992; Antrobus et al., 1995; Stickgold et al., 2001). In particular, the DEs were obtained in up to 50% of cases upon awakening from NREM sleep, especially from stage 2 (Foulkes, 1962; Pivik and Foulkes, 1968; Nielsen, 2000) and REM sleep suppression pharmacologically-induced did not affect dream recall (Landolt et al., 2001; Oudiette et al., 2012).
Although the exclusive relation between REM sleep and dream has been overcome (Scarpelli et al., 2015a), several studies point out that emotional and perceptually vivid contents are more frequent when reported upon awakenings from REM sleep (Foulkes et al., 1988; Nielsen et al., 1991; Merritt et al., 1994; Hobson et al., 2000; Kahn et al., 2002). More in general, REM sleep plays a pivotal role in the processing of emotional events and several studies showed that the consolidation of emotional memories occurs in this sleep stage (e.g., Lara-Carrasco et al., 2009; Nishida et al., 2009). In addition, experimental deprivation of REM sleep has been demonstrated to compromise the consolidation of emotional stimuli (Cartwright et al., 1975; Wagner et al., 2001; Lara-Carrasco et al., 2009; Spoormaker et al., 2014).
Dream contents including negative emotions (e.g., anxiety and fear) are more frequent than positive ones (Foulkes et al., 1988; Nielsen et al., 1991; Merritt et al., 1994; Fosse et al., 2001), and they are often related to waking-life experiences (Stickgold et al., 2001; Wamsley et al., 2010; Eichenlaub et al., 2017; Vallat et al., 2017). In this regard, the existing empirical evidence highlighted that the neural activation of emotional-limbic (Nir and Tononi, 2010) and reward systems (Perogamvros and Schwartz, 2012) during REM-DE contributes to the offline reprocessing of emotions and associative learning (Perogamvros et al., 2013).
Interestingly, clinical studies have also provided evidence on the potential interaction among DE, sleep alterations and affective disorders (Levin and Nielsen, 2007; Nielsen and Levin, 2007; Schredl, 2011). For instance, mood disturbances frequently appear along with sleep modifications involving REM sleep (Benca et al., 1992, 1997; Walker and van der Helm, 2009). More directly, abundant dreams or nightmares are related to REM sleep abnormalities and psychiatric disorders (Cartwright et al., 2003; Modell et al., 2005; Agargun et al., 2007; Schredl et al., 2009; Sjöström et al., 2009; Marinova et al., 2014; Nakajima et al., 2014).
Taking into account the impossibility to directly access to the DE, most studies in the last decades were focused on the retrieval of dream contents upon awakenings (Scarpelli et al., 2015a). Specifically, several investigations using spontaneous or provoked awakenings in the laboratory aimed to find the electrophysiological (EEG) patterns related to dream recall, considering it as a form of episodic-declarative memory (Mangiaruga et al., 2018).
In keeping with some earlier reviews (e.g., Revonsuo, 2000; Revonsuo et al., 2015), it could be hypothesized that DE may play a pivotal role in emotional encoding and regulation, nevertheless the specific function and the neural bases of REM-DEs were directly investigated by very few studies (De Gennaro et al., 2011; Eichenlaub et al., 2018; Vallat et al., 2018a).
In light of the above, here we reviewed the main findings about neural bases of dreaming and the link between EEG correlates of DE and emotional processing, underlying that some evidence on sleep disorders characterized by a peculiar DE, especially occurring in REM sleep, may provide a better understanding on the functional role of dreaming and its potential applications in the clinical field.
Methodological Note
Studies were identified via PUBMED queries. Key search terms included:
- “Dreaming” and “REM sleep” and “Neuroimaging” in title/abstract (62 articles)
- “Dreaming” and “REM sleep” and “Emotional processing” (22 articles);
- “Dreaming” and “EEG” in title/abstract (686 articles);
- “Dreaming” and “PTSD” in title/abstract (450 articles);
- “Dreaming” and “Emotional Memories” in title/abstract (61 articles);
- “Dreaming” and “Narcolepsy” in title/abstract (128 articles);
We grouped the identified articles in the following categories:
1. Cognitive features, emotional aspects and EEG patterns of mental activity during REM sleep;
2. Dreaming and emotional process: (a) role of REM sleep in processing salient emotional waking life experiences; (b) EEG patterns implicated in REM sleep and emotional dreams; (c) associations among traumatic events and nightmares; (d) role of lucid dreams;
3. Dreaming in narcolepsy.
We focused on the role of emotions in REM sleep, excluding non-English articles. All the articles resulting from using this method and related to our focus were included. Following these criteria, we identified 88 publications which were estimated to be of interest for further examination.
Neural Bases of Dreaming and Emotional Processing
A large body of evidence showed that the regions implicated in emotional processes during wakefulness are also responsible for the neurophysiological background of REM sleep that can explain some qualitative features of DE (Maquet et al., 1996; Nir and Tononi, 2010; De Gennaro et al., 2011, 2016; Eichenlaub et al., 2014; Vallat et al., 2018a). In this section, we highlighted the commonalities between the neural bases of REM-DE and emotional processing.
Earlier studies on anatomical correlates of dream alterations by brain lesions studies confirmed the central role of specific cortical and subcortical areas in DE. The main findings point out that two specific cortical systems underlie dreaming. Firstly, a posterior system involves the Temporo-Parieto-Occipital Junction (TPJ) and lesions located in this region alter both waking mental imagery and sleep (Solms, 1997, 2000). The anterior system includes the ventromedial prefrontal cortex (vmPFC) and the white matter surrounding the anterior horns of the lateral ventricles and its damage seems related to dream loss (Solms, 1997, 2000, 2011). It should be noted that many of the afferent and efferent fibers of these two systems are connected with the limbic system and the ventromesial frontal white matter seems to take a part in the interplay between basal forebrain and limbic structures (De Gennaro et al., 2012, 2016).
More recently, human neuroimaging studies have gained a crucial role in dream research. Positron Emission Tomography (PET) and functional Magnetic Resonance Imaging (fMRI) were used especially in the measurement of functional changes in the brain during REM sleep. It has been demonstrated that the distribution of brain activity in REM sleep is not homogenous (Maquet, 2000; Nir and Tononi, 2010). Brain imaging studies found increased regional brain activity in the limbic and paralimbic structures, pontine tegmentum, thalamus and basal forebrain during REM sleep, as compared to wakefulness (Braun et al., 1997, 1998; Nofzinger et al., 1997; Maquet, 2000) and NREM sleep (Braun et al., 1997, 1998; Maquet et al., 2005). In addition, one of the first neuroimaging studies collecting dream report (Maquet et al., 1996) found a bilateral activation of the amygdala in subjects reporting DE upon awakenings from REM sleep, not providing any comparisons with non-recall conditions or awakenings from NREM sleep.
Moreover, remarkable activation of motor and premotor regions has been found in healthy subjects during REM sleep (Braun et al., 1997; Maquet et al., 2000; De Carli et al., 2016), albeit this sleep stage is characterized by muscular atonia thanks to the inhibition of spinal motor neurons by the ponto-bulbar reticular formation (Lai and Siegel, 1999). This seems in line with the observation that individuals affected by REM sleep behavior disorder (RBD) exhibit motor behaviors linked to their DE (Oudiette et al., 2009). Other brain areas are hypoactive during REM sleep compared to the waking state, such as the dorso-lateral PFC (dlPFC), precuneus, orbitofrontal cortex (OFC) and posterior cingulate gyrus (Maquet et al., 1996; Braun et al., 1997; Nofzinger et al., 1997). This evidence could explain the altered executive functions, time perception, and the lack of insight during DE (Desseilles et al., 2011).
Although it is well-known that dreaming is not restricted to REM sleep, some REM sleep features make it a privileged background for DE, especially characterized by high vividness, bizarreness and emotional load (Carr and Solomonova, 2018). In this respect, it should be underlined that most of the regions involved in emotional memory encoding and consolidation (Phelps and LeDoux, 2005; Armony, 2013) are highly activated during REM sleep. In particular, the higher activation of amygdaloid complexes, hippocampal formation and anterior cingulate cortex (ACC) in REM sleep than in wake and/or NREM sleep may explain the emotional intensity of DE reported upon REM sleep awakenings (Maquet et al., 1996; Braun et al., 1997; Nofzinger et al., 1997; Desseilles et al., 2011; Corsi-Cabrera et al., 2016). Some authors suggested that these structures may be responsible for the reprocessing and consolidation of emotional experiences during REM sleep (Hobson and Pace-Schott, 2002; van der Helm et al., 2011; Deliens et al., 2014).
According to the idea that DE play a role in the emotional processing, more recently awake fMRI measures revealed that subjects experiencing frequent fear during DE have a higher activation of mPFC and reduced activation in the insula, amygdala and midcingulate cortex when faced with aversive stimuli (Sterpenich et al., 2019). The results are consistent with the view that mPFC should exert an inhibitory control on fear expression by reducing amygdala activity (Phelps et al., 2004). A large amount of dream reports were collected at home and the presence of fear in DE has been considered as a trait-like factor of the participants (Sterpenich et al., 2019).
These results are consistent with the continuity hypothesis, namely the possibility that dreams and wakefulness shared similar mechanisms (Schredl, 2003, 2009; Scarpelli et al., 2015a,b; Mangiaruga et al., 2018; Sterpenich et al., 2019).
In fact, neuroimaging data in waking state have provided strong support for a crucial role of the aforementioned regions in emotional processing (Phan et al., 2004). It has been proposed that in the waking brain subcortical regions such as amygdala, nucleus accumbens, locus coeruleus, pulvinar, along with the ACC and OFC are involved in conscious and unconscious emotional processing (Morris et al., 1998; Vuilleumier et al., 2001; Kirov et al., 2012). Specifically, the unconscious perception of emotional stimuli has been related with the functional integrity of subcortical areas that receive an executive cortical feedback from the cortical network (e.g., dlPFC, OFC, and posterior cingulate cortex) only during wakefulness (Tamietto and De Gelder, 2010).
It is well-known that amygdala plays a pivotal role in emotional regulation. Several studies found that this structure is responsible for detecting, generating, and maintaining fear-related emotions (for a review see Phan et al., 2004). Moreover, the amygdala is important in the coordination of adequate responses to threat and danger and it has also been demonstrated that is critical also for response to stimuli that predict positive and negative future outcomes (Paton et al., 2006), allowing the organism to learn more about a stimulus and organize adaptive behavior (Maren, 2011). Specifically, the amygdala, thanks to the connections with the hippocampus, thalamus, mPFC, and ACC, is involved in control of the encoding and retrieval of affective memories and the physical expression of emotions during the waking state (Misane et al., 2005). Also, the hippocampus has an established role in emotional memory encoding and retrieval. Along with the amygdala, the hippocampal formation mediates the processing and execution of fear memories (Phelps, 2004).
Structural brain imaging approaches offer the possibility to focus on ultrastructural, anatomical measures and inter-individual variability associated with DE. This perspective allows researchers to overcome one of the intrinsic limitations of the study of dreams, concerning the difficult to exactly define the time-coupling between the sleep stages and the occurrence of DE. Neuroanatomical parameters provide a relatively stable measure to account for some trait-like features of DE. In this vein, the correlations between inter-individual differences in quantitative/qualitative features of DE reported by subjects and structural parameters of limbic areas were observed (De Gennaro et al., 2011, 2016). Once again, the regions engaged in emotional encoding and consolidation showed a strong relationship with dream contents, consistently with the studies revealing that the structural characteristics of hippocampus and amygdala are associated with cognitive and emotional processing in waking tasks (Maguire et al., 2000; Bohbot et al., 2007; Iaria et al., 2008). Specifically, microstructural measures were related to the emotional load, bizarreness and vividness of DE reported. For what concerns emotional intensity, a relation was found between lower structural integrity of the left amygdala and decreased emotional load. Moreover, emotionality was linked to a smaller volume of the left hippocampus and larger volume of the right hippocampus. Also, dream contents characterized by high bizarreness were associated with smaller left amygdala, smaller right hippocampus and lower mean diffusivity of the right amygdala.
These observations were partially replicated on patients with Parkinson's Disease (PD), pointing that PD patients did not differ from healthy subjects with respect to sleep and dream characteristics or the neuroanatomical measures and confirming that vividness is related to the amygdalae and also the thickness of the left mPFC. It should be noted that, along with amygdala, mPFC plays a key role in the acquisition, consolidation and retrieval of fear memory and, specifically, modulates fear learning and extinction (Marek et al., 2013).
Moreover, this study on PD focused on the role of the dopaminergic system in DE. It has been found that higher dopamine agonist dosage was associated with qualitatively impoverished DE, as indicated by reduced bizarreness and emotional load levels (De Gennaro et al., 2016). It is worth noting that one of the main regulator of mesolimbic-dopamine network is the mPFC that makes direct and indirect connections to the amygdala and hippocampus (Patton et al., 2013). In other words, the recent findings on PD confirmed the evidence of an essential interplay between the mPFC and mesolimbic-dopamine system for dreaming, as previously showed by data from patients who underwent prefrontal leukotomy that stopped dreaming (Solms, 2011). In fact, in these patients, the ventromedial white matter containing dopaminergic projections to the frontal lobe were disconnected (Bradley et al., 1958). The importance of mPFC in dream processes has been confirmed in another study using PET (Eichenlaub et al., 2014). They revealed that subjects who recall more frequently their DE (i.e., High Recallers—HR) showed higher regional cerebral blood flow (rCBF) in the mPFC during REM sleep and wakefulness than “Low Recallers” (LR) along with higher rCBF in the TPJ during REM, NREM sleep and wakefulness (Eichenlaub et al., 2014). The structural data obtained by MRI from the same sample of Eichenlaub et al. (2014) and from another research (Vallat et al., 2018b) were analyzed, confirming the relationship between brain anatomical structures and dream recall rate (Vallat et al., 2018a). Specifically, the anatomical measures of the mPFC were significantly different between HR and LR (i.e., increased mPFC white-matter density in HR compared to LR was found). It should be underlined that mPFC is engaged in mind representations and evaluation (for a review, see Legrand and Ruby, 2009), that could have a crucial role in the emotional processing. More in general, it has been hypothesized that mPFC is related to cognitive aspects of emotional processing, such as appraisal/identification of emotion, attention to emotion, awareness and introspection (Phan et al., 2004). Furthermore, it has been demonstrated the involvement of this region in the social emotions processing (Ruby and Decety, 2004). Recently, it has been demonstrated that traumatic experiences and pathological memories are linked to abnormal interactions between hippocampus and mPFC that are responsible to reduced mnemonic ability and decreased emotional control (Maren et al., 2013; Jin and Maren, 2015).
Moreover, it is worth noting that lesion, imaging and stimulation data revealed that the TPJ is related to emotional processing and especially to the “theory of mind,” empathy and social cognition in the waking state (Saxe and Kanwisher, 2003; Young et al., 2010; Santiesteban et al., 2012; Van Overwalle and Vandekerckhove, 2013; Jeurissen et al., 2014). In particular, TPJ contributes to the mentalizing and emotional state inference/attribution (Donaldson et al., 2015; Ye et al., 2015; Biervoye et al., 2016).
Taken together, these findings revealed that, to some extent, the processes that regulate dreaming and emotional salience in DE share similar neural substrates of those controlling emotions during wakefulness. Moreover, in the light of the results by Eichenlaub et al. (2014), one might speculate that the high dream recaller may be particularly interested in their inner world (Ioannides et al., 2009), given that salience seems to have a pivotal role in dream recall, as evaluated in older individuals measuring episodic memory recall (for a review, see Mangiaruga et al., 2018).
(rest in Link)