Dreaming As a Story-Telling Instinct
Total Page：16
File Type：pdf, Size：1020Kb
Dreaming as a story-telling instinct The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters Citation Pace-Schott, Edward F. 2013. Dreaming as a story-telling instinct. Frontiers in Psychology 4:159. Published Version doi:10.3389/fpsyg.2013.00159 Citable link http://nrs.harvard.edu/urn-3:HUL.InstRepos:11181194 Terms of Use This article was downloaded from Harvard University’s DASH repository, and is made available under the terms and conditions applicable to Other Posted Material, as set forth at http:// nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of- use#LAA OPINION ARTICLE published: 0 2 April 2013 doi: 10.3389/fpsyg.2013.00159 Dreaming as a story-telling instinct Edward F. Pace-Schott* Department of Psychiatry, Harvard Medical School and Massachusetts General Hospital, Charlestown, MA, USA *Correspondence: [email protected] Edited by: Jennifer M. Windt, Johannes Gutenberg-University of Mainz, Germany Reviewed by: Jennifer M. Windt, Johannes Gutenberg-University of Mainz, Germany Dreams create new stories out of nothing. tomography (PET) as those regions show- by imagination of future scenarios and Although dreams contain themes, con- ing task-induced deactivation (Gusnard concerns (Schacter et al., 2007; Andrews- cerns, dream figures, objects, etc. that cor- et al., 2001; Raichle et al., 2001). Hanna et al., 2010; Andrews-Hanna, 2012; respond closely to waking life, these are Subsequently, it was discovered that Schacter, 2012). The central nodes acti- only story elements. The story itself weaves temporal synchrony of low frequency vate along with most tasks that recruit one these mnemonic items together in a man- (0.01–0.1 Hz) spontaneous fluctuations of or the other subsystem (Andrews-Hanna, ner far more novel than a simple assem- the blood-oxygen dependent (BOLD) sig- 2012). blage or collage, producing an experience nal of fMRI identifies both anatomical and Synchrony of BOLD fluctuations having a life-like timeframe and life-like functional connectivity among regions among components of the default net- (if often bizarre and impossible) causal- of the default network (Fox and Raichle, work persists into light (Drummond et al., ity (Pace-Schott, 2007; Hobson, 2009). It 2007; Greicius et al., 2008). This network 2005; Horovitz et al., 2008; Larson-Prior is as if one is immersed in another “real- consists of (1) medial parietal areas: pos- et al., 2009)andStage2(Laufs et al., ity” entirely of one’s own non-volitional, terior cingulate (pCC) and retrosplenial 2007) NREM sleep. However, in slow- making (see Rechtschaffen, 1978). Given (Rsp) cortices; (2) posterior-lateral areas: wave sleep (SWS), frontal regions may this phenomenology, it’s not difficult to inferior parietal lobule (IPL), temporo- uncouple from the rest of the default net- see why some indigenous animist soci- parietal junction (TPJ), lateral temporal work (Horovitz et al., 2009; Samann et al., eties believe dreams represent a separate cortex (LTC), temporal poles (TP); (3) 2011,butseeKoike et al., 2011). In the world parallel with waking life (Nielsen, medial temporal regions: hippocampal one study examining REM, unlike both 1991). But neuroscience offers some other formation (HF), parahippocampal cor- waking and NREM, there appeared a lack explanations. tex (PHC); and (4) medial prefrontal of connectivity between the dorsomedial Recent speculations have focused on areas: ventromedial (vmPFC) and dor- prefrontal subsystem and the posterior the brain’s “default network” as a possible somedial (dmPFC) prefrontal cortices central node of the default network in the neural substrate of dreaming (Pace-Schott, (Buckner et al., 2008; Spreng et al., 2009; pCC (Koike et al., 2011). Koike et al. spec- 2007, 2011a,b; Nir and Tononi, 2010; Andrews-Hanna, 2012). ulate that this disconnection contributes Wamsley and Stickgold, 2010; Domhoff, Resting state functional connectivity to the illogic and bizarreness of dream 2011). The default network consists of analyses of BOLD oscillations in wak- cognition, as has also been suggested for regions that, in the absence of exterocep- ing have identified two default-network loss of antero-posterior EEG synchrony tive attention or narrowly focused men- subsystems each of which fluctuates syn- in the fast, gamma (>30 Hz) frequencies tal effort, support self-directed concerns, chronously with central nodes in the pCC during REM (Corsi-Cabrera et al., 2003, immersion in one’s inner life (e.g., day- and anterior medial PFC (amPFC) but 2008). dreaming) or imagining the inner life of notwitheachother(Buckner et al., 2008; Earlier PET and fMRI activational stud- others (Theory of Mind) (Buckner et al., Andrews-Hanna et al., 2010; Andrews- ies of sleep also showed distinctly differ- 2008; Andrews-Hanna, 2012; Buckner, Hanna, 2012). The dorsomedial prefrontal ent activity in medial limbic versus lateral 2012). Most importantly for the current subsystem includes the dmPFC, LTC, TPJ, association cortex during REM sleep. After topic, the default network also simu- and TP whereas the medial temporal lobe sleep onset during NREM, widespread lates future scenarios and re-creates past subsystem includes the HF, PHC, Rsp, cortical and subcortical areas become less ones drawing upon material in episodic, IPL, and vmPFC. The dorsomedial pre- active (Braun et al., 1997; Maquet et al., autobiographical, and semantic memory frontal subsystem selectively activates dur- 1997; Nofzinger et al., 2002; Kaufmann (Schacter et al., 2007; Schacter, 2012). ing experimental tasks involving reflection et al., 2006). However, with the onset Here I will suggest that such constructive on one’s own mental state and that of of REM sleep, midline limbic regions of activities of the brain represent a “hard- others as well as other forms of social cog- the frontal cortex and subcortex reactivate wired” tendency to represent reality in nition (Andrews-Hanna et al., 2010; Mar, to levels equaling and sometimes exceed- the form of narrative—a “story-telling” 2011; Andrews-Hanna, 2012). In contrast, ing those of waking, whereas lateral and instinct or module. the medial temporal lobe subsystem is posterior-medial cortical areas remain in The default network was origi- selectively activated by retrieval of episodic a NREM-like deactivated state (Maquet nally identified using positron emission and autobiographical memories as well as et al., 1996, 2005; Braun et al., 1997, www.frontiersin.org April 2013 | Volume 4 | Article 159 | 1 Pace-Schott Dreams as story instinct 1998; Nofzinger et al., 1997, 2004). This temporal progression of events and of the fact that vmPFC/pmOPFC damage unique pattern of REM activation has been cause and effect in waking, this ten- leads to confabulation whereas its activa- extensively mapped onto dream experi- dency may become more apparent dur- tion accompanies dreaming. Nonetheless, ences (Hobson et al., 2000; Hobson and ing the non-volitional process of dreaming in both phenomena, involuntary gener- Pace-Schott, 2002; Hobson, 2009; Pace- and pathological states such as delirium ation of an organized, fictive narrative Schott, 2011a). NREM dreams may arise or confabulation (Solms, 1997; Hobson, entirely lacking insight suggests that they from more transient activations of similar 1999; Schnider, 2003, 2008; Hirstein, may share neural mechanisms. Possibly, networks (see Nielsen, 2000). Abnormally 2005). the loss of pre-frontally mediated real- intensified dreaming, such as nightmares Like dreaming, spontaneous behavioral ity monitoring, due either to pmOPFC (Levin and Nielsen, 2007), or loss of confabulation involves a fictive narrative damage in confabulation or lateral frontal dreaming (Solms, 1997) may represent, produced effortlessly, without insight as to inactivity during dreaming, may release respectively, a failure to regulate or support its veracity, that is often acted upon by the narrative-production mechanisms from operation of this limbic network. patient (Schnider, 2003, 2008; Hirstein, inhibitory restraint. Therefore, based on activational stud- 2005; Gilboa et al., 2006; Nahum et al., But what about normal narrative ies, the anterior central node (amPFC) 2012). Spontaneous confabulation results production? Braun et al. (2001) per- and much of the mediotemporal subsys- from lesions of the anterior limbic system formed PET conjunction analysis to tem(e.g.,vmPFC,HF,PHC)ofthedefault including posterior medial orbitofrontal identify modality-independent regions network may be similarly active in waking cortex (pmOPFC, part of vmPFC) and its activated by storytelling in both English and REM. However, the posterior key node subcortical connections (Schnider, 2003, and American Sign Language. Although (pCC) and certain components of both 2008; Hirstein, 2005; Gilboa et al., 2006). activity was shared in widespread medial the dorsomedial prefrontal (TPJ, LTL) Confabulated memories usually contain and lateral cortical areas, medial prefrontal and mediotemporal (Rsp, IPL) subsys- real autobiographical events including activation could be most directly related tems may not become similarly active. The those from the remote past (Schnider, to the generic production of narrative dis- one functional connectivity study of REM 2003, 2008; Hirstein, 2005). Such loss course apart from the imagery, lexical, also suggests this disconnection between of insight has been attributed to disrup- and memory processes shared by the two the pCC and dmPFC. Therefore, during tion of