the avatars in the machine

The Avatars in the MachineDreaming as a Simulation of Social Reality

Antti Revonsuo, Jarno Tuominen & Katja Valli

The idea that dreaming is a simulation of the waking world is currently becominga far more widely shared and accepted view among dream researchers. Severalphilosophers, psychologists, and neuroscientists have recently characterizeddreaming in terms of virtual reality, immersive spatiotemporal simulation, or real-istic and useful world simulation. Thus, the conception of dreaming as a simulatedworld now unifies definitions of the basic nature of dreaming within dream andconsciousness research. This novel concept of dreaming has consequently led tothe idea that social interactions in dreams, known to be a universal and abundantfeature of human dream content, can best be characterized as a simulation of hu-man social reality, simulating the social skills, bonds, interactions, and networksthat we engage in during our waking lives. Yet this tempting idea has never be-fore been formulated into a clear and empirically testable theory of dreaming.Here we show that a testable Social Simulation Theory (SST) of dreaming can beformulated, from which empirical predictions can be derived. Some of the predic-tions can gain initial support by relying on already existing data in the literature,but many more remain to be tested by further research. We argue that the SSTshould be tested by directly contrasting its predictions with the major competingtheories on the nature and function of dreaming, such as the Continuity Hypo-thesis (CH) and the Threat Simulation Theory (TST). These three major theoriesof dreaming make differing predictions as to the quality and the quantity of socialsimulations in dreams. We will outline the first steps towards a theory-and-hypo-thesis-driven research program in dream research that treats dreaming as a simu-lated world in general and as a social simulation in particular. By following thisresearch program it will be possible to find out whether dreaming is a relativelyunselective and thus probably non-functional simulation of the waking world (CH),a simulation primarily specialized in the simulation of dangerous and threateningevents that present important challenges for our survival and prosperity (TST), orwhether it is a simulation primarily specialized in training the social skills andbonds most important for us humans as a social species (SST). Whatever the evid-ence for or against the specific theories turn out to be, in any case the conceptionof dreaming as a simulated world has already proved to be a fruitful theoreticalapproach to understanding the nature of dreaming and consciousness.

KeywordsAltered state of consciousness | Avatar | Consciousness | Continuity hypothesis |Dreaming | Evolutionary psychology | Inclusive fitness | Kin selection theory |Need to belong | Practise and preparation hypothesis | Reciprocal altruism the-ory | Simulation | Social brain hypothesis | Social mapping hypothesis | Socialsimulation theory | Sociometer theory | Strengthening hypothesis | The dreamself | The inclusive fitness theory | Threat simulation theory | Virtual reality | Vir-tual reality metaphor

Authors

Antti [email protected] Högskolan i SkövdeSkövde, SwedenTurun yliopisto Turku, Finland

Jarno [email protected] Turun yliopistoTurku, Finland

Katja [email protected] Turun yliopistoTurku, FinlandHögskolan i SkövdeSkövde, Sweden

Commentator

Martin [email protected] Radboud Universiteit Medical CenterNijmegen, Netherlands

Editors

Thomas [email protected] Johannes Gutenberg-UniversitätMainz, Germany

Jennifer M. [email protected] Monash UniversityMelbourne, Australia

Revonsuo, A., Tuominen, J. & Valli, K. (2015). The Avatars in the Machine - Dreaming as a Simulation of Social Reality.In T. Metzinger & J. M. Windt (Eds). Open MIND: 32(T). Frankfurt am Main: MIND Group. doi: 10.15502/9783958570375 1 | 28

http://www.open-mind.net/showAuthor?author=Antti_Revonsuo

http://www.open-mind.net/showAuthor?author=Katja_Valli

http://www.open-mind.net/showAuthor?author=Jarno_Tuominen

http://www.open-mind.net/showAuthor?author=Antti_Revonsuo

http://www.open-mind.net/showAuthor?author=JenniferM_Windt

http://www.open-mind.net/showAuthor?author=Thomas_Metzinger

http://www.open-mind.net/showAuthor?author=Martin_Dresler

http://www.open-mind.net/showAuthor?author=Katja_Valli

http://www.open-mind.net/showAuthor?author=Jarno_Tuominen

http://dx.doi.org/10.15502/9783958570375

http://www.open-mind.net/papers/@@chapters?nr=32

www.open-mind.net

1 Introduction

There may be no Cartesian ghosts residingwithin the machinery of the brain, but still,something rather peculiar is going on in there,especially during the darkest hours of the night.As we sleep and our bodies cease to interact be-haviourally with the surrounding physicalworld, our conscious experiences do not entirelydisappear. On the contrary, during sleep we of-ten find ourselves embodied and immersed in anexperiential reality, an altered state of con-sciousness called dreaming. The Dream Self—the character with which we identify ourselvesin the dream world, and from whose embodiedperspective the dream world is experienced—iswho I am in the dream world (Revonsuo 2005).

But we are not alone in this alternativereality—there are other apparently living, intel-ligent beings present, who seem to share thisreality with us. We see and interact with real-istic human characters in our dreams. Their be-haviour and their very existence in the dreamworld seem to be autonomous. The dreampeople who I encounter within the dream seemto go about their own business: I cannot predictor control what they will say or do. Yet, they,too, are somehow produced by my own dream-ing brain.

On the one hand, dreaming is a solipsisticexperience: when we dream, we dream alone,and outsiders have no way of participating inour dream. Yet on the other hand, dreaming isan intensely social experience, even if the socialcontacts and interactions in the dream worldare merely virtual. In this paper, we will explorethe idea that dreaming is a simulated world, butnot only a simulation of the physical world. It isequally or perhaps even more importantly asimulation of the social world. We will proceedin the following way:

First, we will argue that a remarkable con-vergence has gradually emerged in theoriesabout the nature of dreaming. The field used tobe a disunified battleground of directly oppos-ing views on what dreams are, how exactly theconcept of “dreaming” should be defined, andon the proper level of description and explana-tion for dreaming. Recently, the field has con-

verged towards a more unified understanding ofthe basic nature of dreams. A widely sharedconceptualization of dreaming now depicts it asthe simulation of waking reality. We will brieflydescribe how this theoretical shift has takenplace and where we currently are in the theoret-ical definition of dreaming. This theoretical de-velopment has paved the way for understandingthe social nature of dreams in terms of socialsimulation.

Second, we will explore the nature of so-cial dream simulation in more detail. In whatsense can dreaming be taken as a simulation ofour human social reality? How much and whattypes of social perception and interaction occurin dreams? This question can be broken downinto a number of more detailed questions. Wewill try to answer some of these questions basedon the already existing knowledge and empiricalevidence about the social nature of dreams. Fur-thermore, we will try to formulate more clearlythe questions that cannot yet be answered em-pirically due to the lack of appropriate data.

Third, we will review hypotheses thatalready address the question of the social natureof dreams or assign a social simulation functionfor dreams. Finally, we will outline some basicideas of a Social Simulation Theory (SST) ofdreaming that might offer some explanations forthe social nature of dreams, or at least mightproduce well-defined, testable research questionsconcerning the possible functions of socialdream simulations.

To describe and explain the social natureof dreams as social simulation, concepts bor-rowed from virtual reality technology may beapplied, in this case to the social aspects ofdreaming. One of these concepts is the notion of“avatar”: A simulated virtual human characterwho plays the role of a corresponding real hu-man within a virtual reality. If dreams are vir-tual realities in the brain (Revonsuo 1995), thenwe ourselves within the dream world areavatars, and we interact with other avatars in-side the simulated reality. Somehow, the dream-ing brain is capable of creating credible,autonomous human simulations out of neural


http://dx.doi.org/10.15502/9783958570375


http://www.open-mind.net/

www.open-mind.net

activities in the sleeping brain. A theory ofdreaming as a social simulation should predictwhat kind of avatars are represented in ourdreams, what types of interactions we engage inwith them, and in particular, why it would beuseful to simulate such avatars and interactionsin our dreams—what functions, if any, do theyserve for us.

2 Consciousness as reality-modeling andworld-simulation

Dreaming is the most universal and most regu-larly occurring, as well as a perfectly naturaland physiological (as opposed to pathological),altered state of consciousness. Thus, any plaus-ible (empirical or philosophical) theory of con-sciousness should also describe and explaindreaming as a major state of consciousness.Most theories of consciousness, however, do notconsider dreaming at all or at least do not dis-cuss the results of dream research in any detail(Revonsuo 2006).

Dreaming presents a particularly difficultchallenge for externalist, embodied, and enact-ive types of theories of consciousness.1 They allanchor the existence and nature of conscious-ness to something in the world external to thebrain, or to some kind of brain-world relationsthat, at least partly, reside outside the brain.By contrast, the empirical evidence from dreamresearch shows that full-blown, complex subject-ive experiences similar with or identical to ex-periences during wakefulness (e.g., Rechtschaf-fen & Buchignani 1992), regularly and univer-sally happen during rapid eye movement (REM)sleep. The conscious experiences we have duringdreaming are isolated from behavioural and per-ceptual interactions with the environment,which refutes any theory that states that organ-ism-environment interaction or other externalrelationships are constitutive of the existence ofconsciousness (Revonsuo 2006).

A few theories of consciousness have, how-ever, taken dreaming as a central starting pointin their conceptualization and explanation of

1 The same criticism may to some extent also apply to representationalisttheories of consciousness and dreaming, depending on which externalistor internalist version of representationalism the theory is committed to.

consciousness. When dreaming is taken seri-ously, ideas about the nature of consciousnesstend to converge on internalist theories of con-sciousness that take consciousness and dreamingto be varieties of the same internal phe-nomenon, whose main function is to simulatereality.

One of the earliest attempts to conceptual-ize both waking consciousness and dreaming asthe expressions of the same internally-activatedneural mechanism, only differently stimulated,was put forward by Llinás & Paré in 1991:

[C]onsciousness is an intrinsic propertyarising from the expression of existing dis-positions of the brain to be active in cer-tain ways. It is a close kin to dreaming,where sensory input by constraining theintrinsic functional states specifies, ratherthan informs, the brain of those propertiesof external reality that are important forsurvival. […] That consciousness is gener-ated intrinsically is not difficult to under-stand when one considers the completenessof the sensory representations in ourdreams. (1991, p. 531)

The argument by Llinás & Paré (1991) wasmostly based on considerations of the sharedneurophysiological mechanisms (in the thalamo-cortical system) that could act as the final com-mon path for both dreaming and waking con-sciousness. Binding information together withinthis system intrinsically generates consciousness(“It binds, therefore I am”, Llinás 2001, p. 261);but only during wakefulness is consciousnessmodulated by sensory-perceptual information—in this model, wakefulness can be seen as adream-like state (Llinás & Ribary 1994).

Although the idea that dreaming simu-lates waking consciousness was implicit in thisneuroscientific theory, Llinás & Paré (1991) didnot consider the phenomenology of dreamingand consciousness in any detail. Theoretical ap-proaches characterizing the nature of dreamingas simulation, based on a combination of philo-sophical arguments and empirical facts aboutdreaming, started to emerge during the 1990s.In Revonsuo (1995) the idea was put forward



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

that consciousness in general and dreaming inparticular may best be characterized as a vir-tual reality in the brain, or a model of theworld that places a (virtual) self in the centre ofa (virtual) world. All experiences are virtual inthe sense that they are world-models ratherthan the external physical world somehow dir-ectly apprehended. While the causal chains thatmodulate the virtual reality are different duringwakefulness and dreaming, the virtual world isontologically the same biological phenomenon:the phenomenal level of organization in thebrain (Revonsuo 1995). All experiences are, ac-cording to this view, in their intrinsic phenom-enal character, no different from dreams.

Metzinger (2003) took this line of thoughtfurther and analysed dreams as complex, mul-timodal, sequentially organized models of theworld that satisfy several important constraintsof consciousness. Dreams activate a global modelof the world (globality), they integrate thismodel into a window of presence (presentation-ality), and this model is transparent to the ex-periencing subject, who takes it to be a realworld and not a mere model of the world(transparency) (see also Windt & Metzinger2007).

In Inner Presence Revonsuo (2006) presen-ted a lengthy analysis and defence of the ideathat dreams are internal virtual realities, orworld-simulations, and argued that conscious-ness in general would be best described and ex-plained by treating dreaming as a paradigmaticmodel system for consciousness. The world-sim-ulation contains the virtual self and its sense ofpresence in the centre of the simulation. Thevirtual self is perceptually surrounded by thevirtual place; the virtual place in turn containsmultiple perceptual contents in the form of an-imate and inanimate virtual objects, includinghuman characters. The virtual objects arebound together from phenomenal features likecolor, shape, and motion, but this binding indreams does not always work coherently,thereby resulting in bizarre feature combina-tions and incongruous or discontinuous objectsand persons in dreams (Revonsuo 2006).

Recently, Windt (2010) has formulated adefinition of dreams that stems from similar ba-

sic ideas. Windt’s definition aims to capture theminimal set of phenomenological features thatan experience during sleep should have in orderto count as a “dream” (as opposed to othertypes of sleep mentation). This definition, al-though not explicitly applying the concept of“simulation”, is consistent with the world-simu-lation model of dreaming. According to Windt,dreams are Immersive Spatiotemporal Hallucin-ations (ISTH): there is a sense of spatial andtemporal presence in dreams; there is a hallu-cinatory scene organized around a first-personperspective, and there is a sense of “now”, alongwith temporal duration. The core feature of adream experience is, in Windt’s ISTH, thesense of immersion or presence in a spatiotem-poral frame of reference. Thus, Windt’s ISTH,as well as Metzinger and Revonsuo’s earlierdefinitions, all involve similar ideas of dreams asinvolving an immersive presence of a virtual selfin a virtual, spatiotemporally organized world-model or simulation.

3 Dreaming as simulation: Converging definitions from dream research

Within empirical dream research, definitions ofdreaming have been highly variable and oftenmotivated by underlying theoretical backgroundassumptions held by the theorist. Thus, thepure description of the explanandum, whichshould come first in any scientific inquiry, hasperhaps been biased by a pre-existing theory asto what might count as the explanans—the en-tities, processes, and concepts that are supposedto explain the phenomenon. We will only brieflymention three approaches to defining (and ex-plaining) dreams in the recent history of dreamresearch, where the definition and description ofthe data seem to have been theoretically motiv-ated.

The field of dream research was, in the1970–1990s, a theoretically disunified field. Thedeep disagreements over finding a definition of“dreaming” that would be acceptable across thefield were noted by Nielsen (2000, p. 853)

[T]here is currently no widely accepted orstandardized definition of dreaming.



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

as well as by Hobson et al. (2000, p. 1019):

[…T]here is no clearly agreed upon defini-tion of what a dream is […] and we are noteven close to agreement.

Hobson’s (1988, 1997, 2001) own definition ofdreaming is (or at least was in his earlier writ-ings) a list of some features of dream experi-ence. According to him, a dream is mentationduring sleep that has most of the following fea-tures: hallucination, delusion, narrative struc-ture, hyperemotionality, and bizarreness. Thisdefinition may be (and was) criticized as includ-ing only paradigmatic late-night REM dreamsthat are spontaneously remembered and onwhich our everyday stereotype of what dreamsare like is based. This bias in the definition to-wards REM dreams might be seen to reflect theunderlying theoretical idea or commitment, ob-vious in Hobson’s earlier theories, that dreamphenomenology should be (reductively) ex-plained by referring to the features of REMneurophysiology.

The opposing, cognitive–psychologicalview of the 1980s and 1990s conceptualizeddreaming as a cognitive process that should beexplained at the cognitive–psychological level(Foulkes 1985). References to the neurophysiolo-gical level were unnecessary. In that time and inthe spirit of functionalism and classical cognit-ive science, the cognitive levels of descriptionand explanation were in general seen to be com-pletely independent of implementation levels,such as neurophysiology. Furthermore, dreamingwas thought to occur in every stage of sleep,not only REM sleep, and rather than being fullof bizarreness was mostly a credible replica ofthe waking world. Thus, according to the cog-nitive approach, an explanation of dreamingcannot be based on neurophysiological mechan-isms in general, or for REM sleep on neuro-physiology in particular. The explanationshould be given at cognitive levels rather thanneurobiological ones. Interestingly, it was prob-ably Foulkes (1985) who first characterizeddreams in terms of the idea and the concept ofsimulation. In 1985 he described dreams ascredible world analogs, an organized form of

consciousness that simulates what life is like ina nearly perfect manner.

A third theoretical definition of dreamingcame from clinical dream research, and reflectedthe long and widespread idea in clinical psycho-logy that dreams restore our emotional balanceand have a psychotherapeutic function. Hart-mann formulated this definition of dreamingmost clearly, when he said that “Dreaming, liketherapy, is the making of connections in a safeplace” (1996, p. 13).

During recent years in dream research, theconcept of simulation has become a widely ac-cepted way of characterizing and definingdreaming, as well as a way of formulating theor-etical ideas about the potential functions ofdreaming. Thus, the idea that dreaming is amultimodal, complex, dynamic world-simulationin consciousness during sleep, may be a type ofconception and definition of dreaming thatmany if not most dream researchers are readyto accept (Nielsen 2010). The various contentsof dreams—their events and objects and charac-ters—can be taken to be simulations of theirreal-world counterparts.

Taking Foulkes’s idea of dreams as cred-ible world analogs and as the simulation ofwhat life is like as a starting point for definingdreaming, Revonsuo (1995) formulated the Vir-tual Reality metaphor and later the TST(Threat Simulation Theory) of the evolutionaryfunction of dreaming. This theory is built ontwo background assumptions, the first of whichis precisely the definition of dreaming as “an or-ganized simulation of the perceptual world”(Revonsuo 2000, p. 883). An additional, morespecific assumption of this theory is that dreamexperience is specialized in particular in the sim-ulation of threatening events: it tends to selectand include various types of dangerous enemiesand events and then simulates what it is like toperceive and recognize them (simulation ofthreat perception) as well as how to react andbehaviourally respond to them (simulation ofthreat avoidance behaviours and strategies).Threat simulations appear in a paradigmaticand powerful form especially in nightmares, baddreams, and post-traumatic dreams, but arealso abundant in many other types of dreams



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

such as everyday dreams, recurrent dreams, andin various parasomnias such as RBD (REM-Sleep Behaviour Disorder).

Domhoff (2007), who represents a similarpsychological and content-analysis approach todream research as Foulkes (1985), also charac-terizes dreams as mostly realistic and reason-able simulations of waking life. By emphasizingthat, according to convincing empirical datafrom content-analysis studies of dreams, dreamsimulations are mostly realistic rather thanoverly bizarre and hyperemotional, Domhoff ar-gues against the Hobsonian definition of dream-ing as being full of bizarre contents.

Still, despite their disagreements, bothcamps now seem to accept the notion of simula-tion as a valid description of the core nature ofdreaming. Hobson, in his new protoconscious-ness theory of dreaming and REM sleep (2009),uses the concept of simulation to characterizethe root phenomenon, protoconsciousness, fromwhich both our waking and dreaming conscious-ness arise. According to Hobson, protocon-sciousness is the simulated experiential realityor a virtual reality model of the world that thedeveloping brain turns on during REM sleepeven before birth, to prepare the consciousbrain to simulate the external reality that it willencounter through the senses after birth. Thismodel of the world is genetic, innate, and a hu-man universal. Protoconsciousness acts as thetemplate on which both waking and dreamingconsciousness are built after birth. Thus, ac-cording to this theory, protoconscious dreamconsciousness—a very basic form of an intern-ally simulated world—comes into being prior towaking consciousness, and is causally necessaryfor waking consciousness. As Hobson (2011, p.30) puts it: “I REM, therefore I will be”. Ac-cording to Hobson & Friston (2012), predictivecoding is an underlying mechanism in the brainthat produces predictive simulations of theworld. Therefore, dreaming may also function asa preparatory simulation of the waking world;thus their idea is closely related to the othersimulation-theories of dreaming (Hobson & Fris-ton 2012).

In conclusion, while there still are dis-agreements about many details of dream con-

tent and function, there seems to be relativelywidespread agreement that the definition ofdreaming includes the idea of “simulation” ofthe waking world. The use of the concept of“simulation” to characterize dreaming has re-cently gained wide acceptance in the field. Thesimulation is variously characterized as the sim-ulation of waking life, of waking reality, or ofwaking consciousness, and variously called bydifferent authors a realistic world-simulation, avirtual reality, an immersive spatiotemporalmodel of the world, and so on—but despite thesomewhat varying terminology, the differentterms seem to describe the same basic idea.This conceptual unification is a significant stepforward in the theoretical description and ex-planation of dreams. It paves the way for amore unified theory of dreaming.

4 The simulation of social reality in dreams

Dreaming not only places us into an immersive(virtual) physical reality, but also immerses usinto a (virtual) social reality: in dreams we aresurrounded by close friends and family mem-bers, schoolmates, teachers and students,spouses, romantic partners, old crushes, col-leagues and bosses, celebrities, politicians, ac-quaintances, strangers, and mobs as well asmonsters and other fictitious characters frommovies and video games. All are there in dreamsimulation with us as simulated characters—avatars—and we interact with these avatars inmultiple ways: we perceive, recognize, and se-mantically classify them, we communicate andtalk with them, we collaborate with them, helpthem, criticize them, fight them, escape them,fear them, and love them. At least intuitively,there is no doubt that in our dreams, we liverich and colourful social lives, even if only simu-lated ones.

If dreaming in general can be defined as asimulated world, the question arises whether theconcept of “simulation” can also be usefully ap-plied to describe the social reality of dreams.The first task for a theory that takes theconcept of simulation seriously is to simply de-scribe the social contents of dreams as simula-



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

tions of human social reality. The descriptivequestions can be formulated in more detailalong the following lines:

1. What kind of social perception, social inter-action, and social behaviours are simulated indreams?

2. How frequently are different kinds of socialperception, interaction, and behaviour simu-lated in dreams? How much variation is therein the frequency of different social simula-tions as a function of gender, age, culture,and as a function of the quality and quantityof social interactions during waking life?

It is possible to find answers to many of theabove descriptive questions from the already-ex-isting dream research literature where variousaspects of the social contents of dreams havebeen reported, even if they have not been con-ceptualized as social simulations. In what fol-lows, we will first briefly review some of the ma-jor findings in the literature that describe thequality and the quantity of social simulation indreams. Once we have detailed empirical de-scriptions of the quality and quantity of socialsimulations in dreams, we may seek explanatorytheories and testable hypotheses that could ac-count for why we have social simulation indreams.

4.1 Evidence for simulation of social perception in dreams

From the already existing literature, it is pos-sible to find statistics that describe the qualityand quantity of social simulations in dreams.However, the theoretical concept of “social sim-ulation” is rarely used in dream research literat-ure for interpreting the descriptive results. Here,we will briefly summarize only some of the ma-jor findings.

The minimal criterion for a dream tocount as a social simulation is that the DreamSelf is not alone in the dream but in the pres-ence of at least some other animate character orcharacters. In less than 5% of dreams is thedreamer alone (Domhoff 1996); thus, on thisminimal criterion, dreaming seems to consist-

ently simulate social reality. The other animatecharacters simulated in dreams are predomin-antly human (normative finding in adults isabout 95% human, 5% animal), but the propor-tion of animal characters varies in different cul-tures and age groups, being highest (up to 30–40%) in young children and in adults in hunter-gatherer societies (Domhoff 1996; Revonsuo2000). As human characters are reported in al-most all dreams, and typically there are two tofour non-self characters in a dream (Nielsen &Lara-Carrasco 2007), the presence of simulatedhuman characters must be perceptually detec-ted and registered in the dream by the dreamer.Thus, during dreaming, our neurocognitivemechanisms constantly simulate social percep-tion.

The minimal form of social perception isto detect or register the presence of some humancharacter. A more sophisticated form is the per-ceptual recognition and identification of the hu-man characters who are present, first in termsof some basic perceptual and semantic categor-ies (male/female; familiar/stranger), and thenin terms of more detailed semantic and autobio-graphical information about the precise identityand name of the person. According to the Halland Van de Castle norms, about 90% of simu-lated human characters have sufficiently definitecharacteristics to be semantically categorized,for example as male or female, or as familiar orunfamiliar (Domhoff 1996). Thus, social recog-nition and identification mechanisms are highlyengaged in almost all cases of social perceptionin dreams. The dreamer knows, both during thedream and afterwards when reporting it,whether the simulated characters present in thedream are (or were) male or female, familiar orstrange, friend or family; and in most cases, thefamiliar characters are identified as particularpersons from real life.

Typically, a slight majority of dream char-acters are avatars for familiar persons, althoughthere are well-established gender differences(Domhoff 1996) that might, however, partly de-pend on the gender distribution encountered inthe real-world social environment (Paul &Schredl 2012). In a sample of five hundred REMdreams (Strauch & Meier 1996) familiar people



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

(friends, acquaintances, and relatives) were sim-ulated most frequently (44% of all characters),strangers represented about 25% of dream char-acters, and undefined people about 19%. Inmost dreams, both familiar and unfamiliarpeople were simulated, but in 30% onlystrangers and in 20% only familiar people ap-peared. The mixture of familiar and unfamiliarpeople was true also at the individual level—there were no participants who would have sim-ulated only strangers or only familiar people intheir dreams.

For the most part, the human avatars inthe dream world are quite realistic simulationsof their waking counterparts. The degree ofrealism, however, is difficult to express with ac-curacy by any single measure or quantity, asthere are several features of human charactersthat may independently vary along the dimen-sion of realism (Revonsuo & Tarkko 2002). Theopposite pole for realism is called bizarreness,which in dream research refers to deviationfrom the corresponding entity in waking life.

If any kind and degree of deviation from awaking counterpart is counted as a bizarre fea-ture of a simulated person, then over half of thesimulated humans in dreams (over 60% accord-ing to Kahn et al. 2002; 53% according toRevonsuo & Tarkko 2002) are not perfectlyrealistic simulations. In contrast to other dreamcharacters the Dream Self is rarely distorted inany way (Revonsuo & Salmivalli 1995). Revon-suo & Tarkko (2002) also found that in the vastmajority of cases (around 90% of dream charac-ters), non-self dream characters are perceptuallyentirely realistic—they look the same as theircounterparts look in real life. Where they devi-ate from their counterparts is most often theirverbal and nonverbal behaviour. Thus, althoughthe perceptual simulation of human charactersis nearly flawless in dreams, the simulation ofexpected or predicted behaviours deviate fromwaking norms relatively often, though still atleast a slight majority of behaviours by dreamcharacters are no different from waking life.

Dream characters are also spatially andtemporally quite stable and continuous withinthe dream, although transformations and dis-continuities sometimes do happen (Nielsen &

Lara-Carrasco 2007). A simulated person some-times appears from nowhere, is magically trans-formed into someone else, or suddenly disap-pears without a trace. But these kind of discon-tinuous features account for less than 5% ofdream character features (Revonsuo &Salmivalli 1995; see also Revonsuo & Tarkko2002).

By contrast, the behaviours expressed bydream characters are relatively often to someextent odd or unpredictable. Thus, the simu-lated social reality in dreams is less predictablethan the corresponding social reality duringwakefulness. However, it is unclear how this un-predictability should be interpreted: does itsimply reflect the difficulty (and consequentlyfailure) of simulating complex human beha-viours and interactions realistically by thedreaming brain, or is there some other morefunctional explanation as to why the avatars inour dreams tend to behave in more erratic wayscompared to their waking-life counterparts? Wewill come back to this question when we con-sider the possible functions of social simulationin dreams.

4.2 Evidence for simulation of social interactions in dreams

The Dream Self and other dream characters aresimulated in almost all dreams, but how oftenare they engaged in mutual social interactions?According to Strauch & Meier’s (1996) data (140REM dreams in which a Dream Self was presentand had an active role), in nearly 50% of thesedreams the Dream Self and characters interac-ted, in an additional 20% they acted together,and in 20% they acted independently of eachother. In the rest, the Dream Self acted alone.Thus, social interaction or acting together is typ-ically simulated in dreams where the Dream Selfis present together with some other dream char-acters. When social interaction takes place, thereis almost always verbal communication or con-versation between the Dream Self and the othercharacters, which tends to be focused on con-crete topics (Strauch & Meier 1996), and it isunderstandable and something that would besayable in waking life (Heynick 1993).



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

The more detailed nature of social interac-tions has typically been categorized in terms of“friendly” and “aggressive” interactions.Friendly interactions are on average found inabout 40% of dreams, whereas aggressive inter-actions are somewhat more common, and occurin about 45% of dreams in a normative sample(Domhoff 1996). Strauch and Meier, however,point out that in their sample, neutral interac-tions were also common, and only about half ofthe social interactions in their sample could beclassified as particularly friendly or aggressive.The third category of social interactions thathas typically been quantified in dream reports issexual interactions, but they occur at a very lowfrequency—in Strauch & Meier’s (1996) laborat-ory data, in less than 1% of REM dreams, andin the normative Hall and Van de Castle (Dom-hoff 1996) data, in 4% of women’s and in 12%of men’s dreams collected in a home setting.

In sum, the simulation of dream charactersoccurs very frequently, the characters are per-ceived and recognized by the Dream Self, andthe Dream Self actively participates in commu-nication, social interaction, and joint actionswith the characters. The simulated charactersare also for the most part realistic, stable, andrepresent a variety of different kinds of people.Their behaviours, however, may sometimes beunusual or inappropriate, and not exactly whatwe would have expected from their counterpartsin real life. The tone of the interactions may beneutral, friendly, or aggressive.

When this evidence is taken together, wemay conclude that dreaming simulates a rich,variable, realistic, and concrete, but somewhatunpredictable social reality, inhabited by a mix-ture of familiar, unfamiliar, and undefinedpeople. Therefore, we have solid grounds tostate that dreaming is, among other things, def-initely a social simulation. If this is a universaland ubiquitous feature of dreaming, what kindof theory could explain it? Why does dreamingsimulate social reality at all? It is by no meansself-evident that this should be the case.Dreaming could as well be only a simulation ofsome basic features of the physical world: space,time, objects, events, and the perception of andbodily interaction with the physical world. Or it

could be a simulation of thought processes, athinking-through of our problems, or of ouremotional states and concerns. Moreover, simu-lation of physical objects and their behaviour,or a replay of thinking and emotions, wouldprobably be a simpler task for the brain thanthe simulation of a complex social world. Simu-lation of human bodies and faces and interact-ive behaviours such as conversations seems torequire a lot of energy and computing power—these are very complex phenomena to simulaterealistically. Thus, why does the sleeping brainsimulate social situations in such an intense andinvariant manner? Is there any convincing the-oretical answer to be found to this question?

5 The continuity hypothesis and social simulation theories of dreaming

There are, of course, countless theories ofdreaming. Some have explicitly considered therole of social interactions in dreams, while oth-ers make more general statements about dreamcontent. One of the latter is the Continuity Hy-pothesis (CH), which states that dreams reflectwaking life experiences (Schredl & Hofmann2003) or, more specifically, that our waking con-cerns, thoughts, and experiences have a causalinfluence on subsequent dream content. Thus, ifcertain types of social contacts or interactionsbecome more frequent (or less frequent) in wak-ing life, their simulation in dreams becomes cor-respondingly more (or less) frequent.

This general principle seems to hold inmany cases. For example, in hunter-gatherer so-cieties, where people perceive and interact withwild animals on a daily basis, the proportion ofanimal characters remains high (as it is in chil-dren’s dreams across cultures), whereas inhighly industrialized societies, the animal per-centage decreases dramatically from childhoodto adulthood. But the CH merely restates thisempirical relationship; it cannot answer the the-oretical question of why in young children’sdreams the proportion of animal characters ishigh to begin with. TST (Revonsuo 2000) has at-tempted to answer this question by referringnot to personal experiences in waking life, butto a universal bias that is built into the default



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

values of dream content during human evolu-tionary history.

The CH, even if on the right track inmany cases, is too vague and general as a theor-etical explanation of the details of dream con-tent. It does not predict in any detail how andwhy the causal relationship between waking anddreaming works. It also does not specify in anydetail what counts as a “continuity” and whatwould count as a “discontinuity” between wak-ing life experiences and dream simulations ofthe same. If something happens in waking lifehow closely similar will the dream simulation beto its waking origin, when will the same (or asimilar) content appear in dreams, how fre-quently and for how long will it be incorporatedinto dreams, and so on? These questions havebeen studied under the concepts of day residue(Freud 1950) and the dream lag effect (Nielsen& Powell 1989). The CH takes almost any simil-arity between waking life and dream life as aconfirmation of the continuity hypothesis. But“similarity” as a relationship between two phe-nomena is undefined, ambiguous, and vague.Something that in one respect is similar to itswaking counterpart is in another respect dissim-ilar from it; thus it can be interpreted as eithercontinuous or as discontinuous with waking life.Obviously, if the very same evidence could becounted as either supporting or disconfirming atheory, there is something wrong with how thetheory is formulated.2

As long as the CH remains vaguely formu-lated, almost anything can be counted as its sup-port. If the hypothesis does not specify in any de-tail the potential empirical observations afterwhich its predictions would be falsified, it is notan empirically testable theory. Unless it is formu-lated in a much more specific manner, so thatrisky, exact predictions can be derived from it, itsexplanatory power remains correspondingly weak.In one study where more precise predictions fromCH were derived, the CH was found not to bevalid as a general rule concerning how often dif-ferent everyday activities are reflected in dreams(Schredl & Hofmann 2003).

2 For a recent exchange, see Hobson & Schredl (2011) and relatedcommentaries in the International Journal of Dream Research (2011,vol. 4).

Perhaps a more precise prediction thatcould be derived from CH can be formulated inthe following way: according to CH, dreams rep-resent a random sample of recent waking exper-iences (or a random sample of their memoryrepresentations). The quantities of differenttypes of contents in dreams will therefore pass-ively reflect the proportion of their occurrencein waking life in the recent past (or the memoryrepresentations of waking life). If CH is formu-lated in this manner, as a prediction of randomsampling and passive mirroring of recent wakinglife, then any systematic deviation from a ran-dom sample of waking contents (or memoriesthereof) would count as evidence against theCH. A deviation from passive mirroring of wak-ing life would suggest that some kind of select-ive mechanism is at work. An active selectionbias of particular contents to be either includedin dreams or to be left out would be expectedto result in a disproportionately exaggerated ordiminished frequency of that content in dreamsas compared with waking life. This kind of for-mulation of the predictions of CH makes it atestable theory.

Some more specific suggestions aboutdreaming as social simulation have been putforward in the literature. Brereton’s (2000) So-cial Mapping Hypothesis suggests that dream-ing simulates, among other things, the aware-ness of other persons (social perception) andtheir internal mental states (mentalizing or the-ory of mind-abilities). This theory proceedsfrom an evolutionary standpoint, and considersdreaming as a rehearsal ground for emotionaland perceptual abilities related to the mappingof the body image of the self into an emotion-ally-salient social space. Others have also hypo-thesized that our mindreading abilities couldpotentially be a target of simulated social per-ception in dreams (Kahn & Hobson 2005; Mc-Namara et al. 2007). Moreover, Nielsen & Ger-main (2000) have suggested that dreamingmight simulate attachment relationships and in-terpersonal bonds in ways that would maintaintheir adaptive significance even today, andHumphrey (2000) has compared the social func-tions of dreaming to those of play. The possibil-ity that dreaming simulates pro-social and ag-



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

gressive social interactions in distinct sleepstages, and that these simulations might exert aregulatory influence on our waking social lives,was put forward by McNamara et al. (2005).Last, Franklin & Zyphur (2005) have consideredhow the simulation function of dreams might beexpanded to cover social cognition and complexsocio-cultural situations.3

The problem with the above social simula-tion theories of dreaming is that either they arenot detailed enough to be testable, or that few,if any, have ever been directly tested againstcompeting theories. They are interesting generalideas, but not strictly formulated theories thatcould be directly tested, or from which detailedpredictions and potential explanations for thesocial contents of dreaming could be derived.Thus, these theoretical ideas have not led to astrong empirical, hypothesis-driven researchprogram that would be able to systematicallytest the plausibility of these theories.

Whenever we formulate theories of dream-ing, or of the functions of dreaming, they shouldbe formulated in such detail that empiricallytestable predictions can be derived from them.Statements that are too vague or too general(e.g., “dreams are continuous with waking life”;“dreams are social simulations”) are difficult totest as such. The predictions derived from gen-eral statements are too unspecific. Thus, thetheories remain uninformative but of courseconsistent with almost anything we might real-istically expect to find in dream content. If atheory makes no detailed, risky predictionsabout what should or should not be found indream content (under some specific circum-stances or in specific populations) it doesn’thave much explanatory power, either. So farthere is no detailed, convincing, testable theory

3 Another popular theory of dreaming postulates that the realisticsimulation of character–self interactions serves the function of emo-tion regulation during dreaming (Nielsen & Lara-Carrasco 2007). Inthis group of theories, the function of dreaming is proposed to be thecalming down of emotional surges, such as we see in psychotherapy(Hartmann 1995, 1996, 1998), or as reflecting the extinguishing offear memories (Nielsen & Levin 2007). It is increasingly apparentthat sleep plays a role in the consolidation of emotional memories,but whether sleep also regulates the emotional charge and valence ofmemories is not yet entirely clear (for a recent review, see Deliens etal. 2014). Thus, whether the emotional regulation theory has specificimplications or predictions for social simulations in dreaming is notevident.

of the nature and the function(s) of social simu-lations during dreaming. There is also a lack ofdata on the detailed quantity and quality ofsimulated social interactions in dreams, andhow they relate to real social interactions in thewaking life of the same person. In the rest ofthis paper, we will try to outline ideas for thetheoretical basis of a social simulation theory ofdreaming and to formulate some empirically-testable hypotheses directly derived from thetheory.

6 Towards a testable social simulation theory of dreaming

The relatively loose idea or the general observa-tion that dreams are social simulations needs tobe turned into a theory from which testablepredictions can be derived. There are severalways in which this could be done. In the rest ofthis paper, we will formulate some suggestionstowards that end. The basic assumptions thatwe adopt are based on the earlier work on thedefinition of dreaming (and consciousness) as aninternal world-simulation in general (Revonsuo2006). Any plausible theory of social simulationshould also take into consideration, and drawfrom, concepts and advances in the fields of so-cial psychology and evolutionary biology, in or-der to create a credible theoretical context intowhich social simulations in dreams can beplaced. We will therefore connect the idea thatdreaming may function as a platform for simu-lating social perception and interactions tosome influential evolutionary biological and so-cial psychological theories, as well as to theearlier simulation theory of the original evolu-tionary function of dreaming, the TST (Revon-suo 2000).

The two generally-accepted theories inevolutionary biology that seem to be relevantfor the formulation of an evolutionary SST ofdreaming are the Inclusive Fitness and Kin Se-lection Theory (Hamilton 1964) and ReciprocalAltruism Theory (Trivers 1971). Both are gen-eral evolutionary biological theories that applynot only to humans, but to multiple other spe-cies as well. Further, both have received ampleempirical support from animal and human stud-



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

ies, and could thus serve as solid ground inguiding our thinking about social behaviours inevolutionary biological terms.

The Inclusive Fitness Theory (Hamilton1964) postulates that an individual’s genetic re-productive success is the sum of that indi-vidual’s direct reproduction and the reproduc-tion of the individuals carrying identical genealleles. An individual can improve its overall ge-netic success by engaging in altruistic social be-haviour that is directed towards individuals car-rying identical alleles. The Kin Selection Theoryis a more specific form of the inclusive fitnesstheory, which requires that the shared allelesare identical by descent. Thus, Kin SelectionTheory postulates that an individual can in-crease its inclusive fitness by directing acts ofaltruism specifically towards genetic relatives,whereas inclusive fitness as such is not limitedonly to cases where kin are involved. Both, how-ever, predict that acts of altruism should moreoften be directed towards individuals who shareidentical alleles.

Reciprocal Altruism (Trivers 1971) isdefined as behaviour whereby an individual actsin such a way that temporarily reduces its fit-ness while increasing another individual’s fit-ness. However, individuals engage in altruisticbehaviour with the expectation that the recipi-ent of the altruistic act will act in a similarmanner at a later time. A strategy of mutualcooperation may be favoured when there are re-peated encounters between the same individu-als. Although cheating might be more beneficialfor the individual in terms of immediate re-wards, co-operation might provide net gaincompared to short-term benefits.

Since selection pressures act on the typicalconditions present in the history of any species,consideration of the demographics of the typicalevolutionary environment of humans is crucialfor understanding the evolution of social beha-viours in our species. Recently, Hill et al. (2011)analyzed co-residence patterns among thirty-two present-day foraging societies, assumingthat these might reflect an ancestral humangroup structure. They found that primary anddistant kin of an adult individual accounted forapproximately 25% of the co-resident adult

members of a band, i.e., about 25% of adultmembers in the group were directly geneticallyrelated, whereas about half of the adults wererelated through spouse or siblings’ spouses, andthe other 25% of adults were genetically unre-lated.

If we accept the assumption that this ob-served distribution of relatedness approximatesthe degree of relatedness in ancestral humanbands, there have been ample opportunities forancestral humans to be subjected to selectionpressures that could be explained usingstrategies postulated by the inclusive fitness andKin Selection Theory, as well as Reciprocal Al-truism Theory. There is ample evidence thatpeople are more likely to help their relativesthan genetically unrelated individuals (e.g.,Burnstein et al. 1994), and that lethal violenceis more frequently directed towards genetically-unrelated individuals than relatives (Daly &Wilson 1988). People also tend to be more al-truistic towards other people in single roundprisoner’s dilemma game than could be expec-ted (Frank et al. 1993) in order to protect theirreputations. This seems to be a reasonablecourse of action, given that the faces of indi-viduals labelled as untrustworthy cheaters arebetter recalled than those labelled as cooperat-ive (Mealey et al. 1996). There are also ratherlarge interindividual differences in altruistic be-haviour, depending on factors such as age, sex,tendency to empathize, and circumstantial con-ditions.

The social environment has afflictedstrong selection pressures on human cognitivefaculties, and there are several theories thatconsider our essentially social nature. Dunbar(1992, 2008) has forwarded the Social Brain Hy-pothesis, which states that the main factor inthe increase of our neocortical volume has beenthe cognitive demand bestowed on us by the in-crease in hominid group size. Sutcliffe et al.(2012) propose the idea that the costs and be-nefits of social interactions have been a criticaldriver for cognitive evolution. While our mostintimate relationships are a source of social sup-port, they are also the most costly as the qual-ity of these relationships is dependent on thetime invested in creating and maintaining them



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

over time. Forming weaker and less time-con-suming ties with acquaintances can provide be-nefits such as information exchange and accessto resources without exhausting an individual’sresources that are allocated for social interac-tion. Our individual social worlds thus consistof hierarchically-layered sets of relationshipsdefined by relationship intimacy, and differentrelationship types are designed to have differentkinds of functions.

Turning our attention to the potentiallyrelevant literature in social psychology, somefurther concepts and measures might be con-sidered useful for dream theory. When it comesto the simulation of social interaction, one ofthe most relevant concepts is the social “Needto Belong” (Baumeister & Leary 1995). Thisfundamental motive towards interpersonal at-tachment and close, supportive social bondspervades and influences our actions, emotions,and cognitions, and is fulfilled only by social af-filiation and acceptance. To help us navigatethe complex social world, and attune us to so-cially relevant information, two further advance-ments have been hypothesized in the form ofthe Sociometer Theory (Leary et al. 1995) andthe social monitoring system (Gardner et al.2000). Sociometer Theory proposes an internalmonitoring device that feeds forward informa-tion about our level of social inclusion in theform of self-esteem or self-worth (Leary et al.1998), whereas the social monitoring system ispurported to guide the processing of social in-formation whenever people’s needs to belongare not being met (Pickett et al. 2004). In sum,the concept of “Need to Belong” in general, andthe suggested social monitoring systems in par-ticular, might prove useful in postulating test-able hypotheses for the functions of social simu-lation in dreams. The Sociometer, for example,might act in a similar fashion to the threat cuespostulated in TST, and prompt dreams to simu-late relevant social skills or interactions.

An interesting developmental suggestionabout the interplay between simulation mech-anisms and social deficits has recently beenput forward by Oberman & Ramachandran(2007), who propose that in typically develop-ing individuals the abilities of Theory-of-Mind

(ToM), empathy, perceptual recognition, andmotor mimicry might be mediated by an in-ternal simulation mechanism or mechanisms.By taking into consideration a condition—aut-ism—where all these abilities appear to be im-paired, they make the case for a possible linkbetween deficient simulation mechanisms andbehavioural and social deficits. The exact im-plications of this idea for the hypothesis thatdreams serve a social simulation function re-quires further consideration. One possibility isto test whether individuals with Autism Spec-trum Disorders (ASD) dream less of social in-teractions, or whether their dreams of socialinteractions are different in content from thoseof other people. Thus far this line of researchhas not been explored in depth. Daoust et al.(2008) have looked into the dream contents ofpeople with ASD, and found that they reportsignificantly less dream-characters and socialinteractions than the control group. Theynote, however, possible error sources in thetesting procedure, such as, for example, howthe reporting of dreams itself might be af-fected by ASD.

There has been some research linking theeffects of attachment relationships to dreaming.If, as attachment theory proposes, we use ourearly experiences with primary caregivers andother attachment figures as model states for fu-ture social interactions and the way we viewand attune to our social world, it could be as-sumed that this would also affect our simula-tions of this world. Early attachment and bond-ing are, after all, quintessential for our species,and according to Fonagy & Target (1997) mightalso work as the basis for our abilities to men-talize or to create a ToM. McNamara (1996)has developed the idea that REM sleep is themechanism that activates and maintains earlyattachment relations, as well as pair-bonding inlater life. Selterman & Drigotas (2009) havefound that attachment style is correlated todream emotions when dreaming about romanticpartners, so that those with anxious or avoidantattachment styles reported more stress, conflict,and negative emotions.

In an exploratory study on the dream con-tents of those suffering from Complicated Grief



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

(CG) after the loss of an attachment figure,Germain et al. (2013) found the dreams con-taining family members to become significantlymore frequent, while there was no marked in-crease in the occurrence of deceased characters.Males suffering from CG also reported more fa-miliar persons in their dreams than the controlgroup. Both male and female CG patients alsoexhibited fewer negative emotions and fewer in-stances of aggression in their dreams, and fe-males also had decreased amounts of positiveemotions and friendliness.

We can thus conclude that the inherentlysocial nature of our species is deeply ingrained,and has likely been as important for our sur-vival in the ancestral environment as threat per-ception and avoidance skills. SST can thereforebe formulated in an analogous manner to TST,but in addition to the evolutionary backgroundtheory, also taking into consideration importantsocial functions such as the need to belong, so-cial bonding, social networking, and social sup-port as essential ingredients.

TST (Revonsuo 2000) places the contentsand the function of dreaming in an evolution-ary-psychological context and proposes thatdreams were selected for their ability andpropensity to simulate threatening events in asafe way, thus preparing the individual to sur-vive real-life dangers. The hypotheses and pre-dictions of the TST, especially concerning theinclusion of threat simulations in dream con-tent, have gained support from several inde-pendent sources, such as studies on the contentof nightmares and bad dreams (e.g., Robert &Zadra 2014), recurrent dreams (Valli & Revon-suo 2006; Zadra et al. 2006), post-traumaticdreams in children and adults (Bulkeley & Ka-han 2008; Valli et al. 2006), dreams anticipatinga stressful experience (Arnulf et al. 2014), chil-dren’s earliest dreams (Bulkeley et al. 2005),dreams and mental contents in parasomnias(Uguccioni et al. 2013), the dreams and night-mares of new mothers (which mostly depict theinfant in peril and trigger protective behaviours,Lara-Carrasco et al. 2013, 2014; Nielsen & Lara-Carrasco 2007), as well as dreams of the generalpopulation (for a review, Valli & Revonsuo2009).

Thus, when it comes to emotionally negat-ively-charged dream contents that simulatesome sort of dangerous situation or unfortunateevent, the TST seems able to quite well predictand explain many features of the quantity andthe quality of the threat simulations found inthe data. Therefore, a similar theoretical ap-proach might also prove fruitful in the case ofsocial simulation theory. The SST, however,needs to be formulated in such a manner thatits predictions can be clearly distinguished fromthose of the TST.

As negative and threatening events com-monly occur in dreams, the TST alone alreadycovers a fairly large proportion of dream con-tent. But it also ignores a relatively large pro-portion of dream content, as it does not offerany explanation of non-threatening dreams orfor the simulation of neutral and positive eventsin dreams. This raises the question: do types ofdream events other than those that are threat-ening have some evolutionarily-based simulationfunction, independent of the threat-simulationfunction of dreaming? Are there events that areequally important targets for simulation as thenegative, threatening situations simulated inthreat simulation dreams?

TST covers threatening events in dreams,whether social in nature or not. Many threaten-ing events of course do involve social interaction(such as verbal or physical aggression), but areexplained by the TST as primarily simulationsof specific types of threat, and therefore as re-hearsals of threat perception and threat-avoid-ance behaviours, rather than as simulations ofsocial interactions as such. A social simulationtheory that explains dreams that TST does notcover should thus focus on social simulationsthat are largely independent of the threat-simu-lation function. In some dreams these two typesof simulation may, however, be difficult to teaseapart. For example, a social simulation theorymight account for some social interactions thathappen during a threatening event in a dream,such as how the Dream Self interacts with oth-ers and collaborates with them during a threat-ening situation. Furthermore, these two simula-tion theories may not be mutually exclusive butinstead complement each other. Some specific



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

types of simulations of negative social interac-tions are better accounted for by the TST whileother, positively toned simulations can be ex-plained by the SST. For example, from an evol-utionary perspective it might make sense tosimulate different kinds of interactions, friendlyor aggressive, with people belonging to differentlayers of our social hierarchy.

We are open to the possibility that socialsimulation is an original evolutionary functionof dreams alongside the threat-simulation func-tion of dreaming. We believe that social simula-tion theories hold much promise. But beforethis belief can be empirically justified, a test-able version of the social simulation theoryneeds to be formulated. Such a theory shouldindependently cover the social simulations indreams that fall outside the scope of the TST.

Furthermore, also the predictions of theCH must be distinguished and separated fromthose of the SST. Therefore, the question be-comes: What aspects of human social realitymight dreams be specialized in simulating insuch a way that these social simulations havesignificant consequences for cognition and beha-viour during the waking state, and in virtue ofwhich social simulations during dreaming havefulfilled important functions in the evolutionaryhistory of the human species? What kind of so-cial-cognitive processes and behavioural socialskills might have been both critical enough bothfor an individual’s survival and successful repro-duction, as well as occurring frequently and uni-versally enough in the human ancestral environ-ment, to be selected for as a universal feature ofhuman dreaming? Moreover, those processesand skills would have to be something that infact can be regularly simulated by the dreamingbrain, and they have to be contents that actu-ally are being simulated frequently and univer-sally in human dreaming, according to the evid-ence from content analysis studies of dreaming.

To sum up, a credible version of the SSTshould have predictions and explanations thatare clearly different from both the TST and theCH. To be different from TST, the SST shouldpredict and explain the social simulations thathappen outside threatening events in dreams,and to be different from the CH, the SST

should predict that some types of social stimuli,social cognition, or social behaviours are simu-lated actively and selectively, so that they areoverrepresented in dreams as compared to wak-ing life.

We will first consider some basic cognitiveprocesses that might fulfil these roles and willthen proceed to more complex social behavioursand interactions. We admit that many of theseideas are at this stage speculative. But if it ispossible to formulate them in an empiricallytestable manner, then we can figure out later onwhich ideas remain mere empirically unsuppor-ted speculations, and which ones might actuallypredict and explain central aspects of our dreamcontent.

6.1 The simulation of social perception asa function of dreaming

Overall, there are good reasons to support theview that fast and errorless social perceptionabilities were universally important skills for hu-mans during their evolutionary history, and,therefore, rehearsing them through dream simu-lations would have served to maintain and en-hance their speed and accuracy during wakeful-ness. In the ancestral environment, fast and effi-cient social perception and recognition mechan-isms were essential for telling friends and alliesapart from potential enemies. Thus, detectingthe presence of other human beings in the samespatiotemporal context where oneself is located,immediately classifying them in terms of famili-arity, identity, and history of past interactionswith them, and predicting the nature of futureencounters with them must have been an im-portant survival skill. Perhaps it was importantenough that rehearsal of these social-cognitivefunctions through social simulations duringdreaming would have increased an individual’sinclusive fitness.

The social perception system needs toquickly estimate answers to the following ques-tions: am I alone in here or are there other hu-mans present? Are the other humans around mefamiliar to me or are they strangers? Thus, thefirst stage of social perception is to detect otherhumans in the vicinity and to classify them in



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

terms of unfamiliar people (strangers) vs. famil-iar people. As Diamond (2012) explains in “TheWorld Until Yesterday”, in most traditional so-cieties during human evolutionary history, toencounter strangers was unusual and typicallyconsidered potentially dangerous, because thesocial interaction that followed might not neces-sarily have been peaceful in nature.

The second stage of social perception dealsin more detail with the familiar people that aredetected. If the people in my presence are famil-iar to me, who exactly are they? What is my re-lationship with them? What have my past inter-actions with them been like? What should I ex-pect the interaction between us to be like thistime around? To answer these questions, famil-iar people need to be quickly identified. Basedon semantic and autobiographical memory in-formation that we have about people familiar tous, we quickly activate expectations andstrategies as to how we should interact with thepeople around us in the most constructive way.

But so far this idea is mere speculation.What kind of testable hypotheses and predic-tions could be derived from this theory? Howcould we derive predictions that clearly distin-guish the SST from the CH? The CH does notattribute any evolutionary simulation functionsto dream content; according to CH, dreamingsimply and passively mirrors whatever experi-ences have recently been encountered in thedreamer’s waking life (and thus impressed onlong-term memory). Obviously, therefore, itwould not lend sufficient (or specific) support tothe SST to predict that social perception shouldbe found in dreams in the same proportions asin waking life, because the CH predicts and ex-plains exactly the same observation and,moreover, does it more parsimoniously, withoutpostulating any just-so-story of evolutionaryfunctions to social dream content.

The SST must thus go beyond the CH andmake the risky prediction that, if social percep-tion is the original evolutionary function ofdreaming and it is therefore still expressed inour dream contents, then dreams are specializedin simulating social perception. If dreams arespecialized in simulating social perception, thenperceptual contents, cognitive processes, and

behaviours relating to social perception skillsshould occur (as simulations) in a selective orexaggerated form in our dreams. The testableprediction derived from this is that duringdreaming, social perception occurs more fre-quently than in waking life (shows quantitat-ively an increased frequency) and/or qualitat-ively in a more difficult or challenging formthan in waking life.

Quantitatively, dream simulations couldexaggerate the proportion of the types of stim-uli that were most important to recognizequickly and accurately during evolutionary his-tory (e.g., strangers vs. familiar people; enemiesvs. friends). It is important to process this in-formation quickly because the information hadhigh survival value in ancestral environments.Furthermore, dream simulations could presentqualitatively challenging stimuli for the socialperception system; for example, more variety ofdifferent kinds of stimuli (different kinds of fa-miliar and unfamiliar simulated people), or am-biguous stimuli that are more difficult to per-ceive or interpret than real life stimuli (vague orunstable simulations of people).

Conversely, if the social stimuli in dreamssimply mirror the social stimuli during wakeful-ness (and memory representations of them),quantitatively and qualitatively, then the CHgains support: dream experiences merely copythe patterns and rates of social stimulation en-countered during wakefulness, but do not select-ively and actively simulate them in ways andproportions that would reflect some originalevolutionary functions and would therefore havesupported important survival skills in ancestralenvironments.

To test these two opposing theories, SSTand CH, against each other empirically, we needdetailed information not only about the quantityand quality of social perception in dreams, butalso about the quantity and quality of social per-ception during wakefulness in the same subjects’lives during the same period of their lives. Somestudies already exist that provide us with thiskind of data, but most of the hypotheses remainto be tested in future studies that should be ex-plicitly designed to test the opposing hypothesesand predictions of the two theories.



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

McNamara et al. (2005) conducted an in-teresting study that can be interpreted as test-ing the SST prediction that social perception isquantitatively exaggerated in dreams as com-pared to waking life. They conducted experiencesampling from fifteen individuals over twoweeks across waking, REM sleep, and Non-Rapid Eye-Movement (NREM) sleep states.The participants recorded verbal reports oftheir perceptual and other experiences whenpaged at random intervals during sleep or wake-fulness.

The results showed that more charactersappeared in dreams than in wake reports. Unfor-tunately McNamara et al. (2005) do not reportthe exact descriptive statistics of this finding, sowe do not know how large this difference ex-actly was. In any case, this finding is better inaccordance with the predictions of the SSTthan CH: Stimuli requiring social perception(human characters) are present at higher fre-quencies during dreaming than during wakeful-ness, when experiences from both states aresampled and reported in a similar manner.

This important finding suggests that thebasic processes and skills required in social per-ception are more engaged during dreaming thanduring an equal stretch of time in wakefulness.This lends support to the hypothesis thatdreaming is specialized in the simulation and re-hearsal of social perception, which may thus beone of the original evolutionary functions ofdreaming. It has to be added, however, thatMcNamara et al. (2005) is the only study so farthat provides us with this kind of data, wherethe frequencies of the social contents of dream-ing and waking experiences have been directlycompared with each other. Replications are ob-viously required in different populations and inlarger samples of dreams and waking experi-ences. But so far, so good for SST.

The same study can be taken to test theadditional prediction of SST, namely thatdream simulations of human characters shouldexaggerate the proportion of the particulartypes of stimuli that were, during evolutionaryhistory, most important to recognize quickly.Meeting strangers posed a threat in the originalevolutionary context; thus, the SST predicts

that strangers or unfamiliar people should beoverrepresented in dreams as compared to wak-ing life, to simulate and rehearse the type ofperceptual categorization (familiar vs. unfamil-iar) that was most important in the evolution-ary context. McNamara et al. (2005) reportthat the proportion of strangers (or unfamiliarpeople) encountered in dreams is indeed signi-ficantly higher than in waking life. Only 25% ofpeople present in the waking episodes were un-familiar, whereas about 50% of the (simulated)people in dreams were unfamiliar. Again, thisdiscrepant pattern is well predicted by and ac-counted for by the SST, but goes against thepredictions of the CH.

The recognition and identification of famil-iar people as who exactly they are could alsopotentially be a target of useful simulation indreams. It might be argued from SST thatquick and correct recognition of familiar peopleenhances the quick selection of the appropriatesocial strategies and behaviours when we inter-act with them. As about 50% of simulatedpeople in dreams are familiar, there are stillplenty of opportunities to rehearse these recog-nition skills. There are, however, no studies thatwould have directly and quantitatively com-pared the frequency of face recognition duringdreaming and wakefulness. But still, there aresome studies that question whether face recog-nition is engaged during dreaming and to whatextent.

Kahn et al. (2002) report, in a characterrecognition study, that about 45% of familiardream characters were recognized through theirappearance (including facial features), and anadditional 12% by their observable behaviour.Thus, nearly 60% of dream characters are recog-nized perceptually. However, about another 12%of dream characters are recognized intuitively,by “just knowing” who they are, which suggeststhat in those cases, the “recognition” happensin a top-down manner and is therefore inde-pendent of the perceptual and facial features ofthe dream character.

If familiar persons are not overrepresentedin dreams to begin with (as the McNamara etal. 2005 study suggests), and only well under50% of the familiar people simulated in dreams



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

are recognized through their facial features, thispattern of data does not particularly supportthe idea that dreams are specialized in rehears-ing familiar face recognition. However, we stilllack knowledge about the frequency of face re-cognition in waking vs. dreaming, and only astudy directly making that comparison couldproperly test this idea. So, the case remainsopen, but the expectations are not particularlyhigh that this prediction of the SST will gainstrong support in the future.

6.2 The simulation of mindreading as a function of dreaming

In addition to the processing of familiarity andidentity, another aspect of social perception iscalled Theory-of-Mind (ToM) or “mindreading”.This refers to the interpretations we automatic-ally make about the internal mental states ofthe people around us. We not only categorizethe people around us as familiar and unfamiliar,and assign an identity to familiar persons, wealso attribute thoughts, beliefs, motives, andemotions to them. As mindreading is crucial forour ability to predict and explain other people’sbehaviours, our mindreading abilities could po-tentially have been a target of simulation duringsimulated social perception in dreams (Kahn &Hobson 2005; McNamara et al. 2007).

The study by Kahn & Hobson (2005)quantifies the frequency of mindreading activit-ies in dreams. In one sample of thirty-five parti-cipants and about nine dream reports per parti-cipant, about four dream characters per reportwere observed on average. In over 80% of thesedreams, the participants reported having hadengaged in mindreading (at least one of) theother dream characters’ internal mental states.In another sample, 24 subjects reported on av-erage six dreams per participant. Each dreamwas divided into separate dream events (on av-erage four events per report were found), andthe participants were asked to report, concern-ing each event, whether or not they were en-gaged in mindreading the other dream charac-ters. In 50% of the episodes, mindreading wasreported to have occurred. Thus, on the basis ofthese results, we may say that mindreading fre-

quently occurs during dreaming. Kahn & Hob-son (2005) in fact suggest that this may beevidence for a specific simulation function beingat work:

The two studies undertaken here supportthe idea that dreaming may provide a sim-ulation of waking life as suggested byRevonsuo (2000), though not restricted toonly threatening events. Instead, the dataof these studies suggest that if dreaming isa simulation process, it is a simulationthat provides a way of knowing and deal-ing with the intentions of others, both pos-itive and negative. (p. 56)

The above studies show that mindreading iswell represented in dreams, but they cannot tellus whether mindreading is overrepresented indreams, as its frequency of occurrence cannotbe directly compared to waking life. However,McNamara et al. (2007) have conducted a directcomparison of the frequency of mindreadingbetween waking experiences, REM dreams, andNREM dreams of the same subjects. This iswhat they found:

REM reports were three times as likely tocontain instances of mind-reading as werewake reports and 1.3 times as likely asNREM reports. Of 100 reports per state,there were 39 instances of mind-reading inREM reports, 29 in NREM reports, and12 in wake reports. (McNamara et al.2007, p. 211)

In conclusion, from looking at these studies, wemay say that mindreading activities frequentlyoccur in dreams, and that their frequency of oc-currence is significantly greater during dreamingthan during wakefulness: Mindreading isoverrepresented or exaggerated during dream-ing. Thus, this data supports the SST predictionthat dreaming specifically simulates mindreadingin order to maintain and rehearse ourmindreading abilities, rather than the CH pre-diction that dreaming simply reflects theamount of mindreading we engage in duringwake experiences.



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

Another finding that might indirectlylend support to the SST-mindreading idea isthat the behaviours and communications ofdream characters are often bizarre (Kahn etal. 2002; Revonsuo & Salmivalli 1995; Revon-suo & Tarkko 2002); that is, they are unusual,unexpected, and thus unpredictable on thebasis of our waking expectations. Studies onintentional social interactions between theDream Self and other avatars in lucid dream-ing suggest that dream characters are largelyindependent of the dreamer and behaveautonomously (Stumbrys et al. 2011; Tholey1989). Unusual and unpredictable behaviourscould be interpreted simply as failures of thedream simulation to produce credible se-quences of real-life behaviour. But they couldalso be interpreted as particularly engagingand activating social stimuli that serve tochallenge our mindreading skills. That is,bizarreness in this case could be functional inthe sense that it makes the simulation morechallenging. Perception of unexpected beha-viours may trigger a reconsideration of what isgoing on in the character’s mind in order toproduce such unexpected behaviour, and thuspresent a frequent need to engage inmindreading as we interact with unpredictablecharacters in our dreams. This idea could beempirically tested by studying whether bizarrebehaviours on the part of dream characterstend to trigger mindreading in the DreamSelf, and whether this feature of dreams mightpartially explain the apparently frequent en-gagement in mindreading in dreams.

6.3 The simulation of social interactions as a function of dreaming

Humans are an essentially social species and anindividual’s survival in the ancestral environ-ment was most likely entirely dependent on theindividual’s ability to form long-lasting positivesocial bonds with close kin and other groupmembers who offered protection, access to nu-trition and other crucial resources for survival,collaboration, friendship, social support, matingopportunities, and opportunities to gain a bet-ter social status within the group.

Social interaction in dreams is a morecomplex affair than simple social perception.There need to be some behaviours that linkdream characters and the Dream Self, where theintentional behaviour of one character (or theDream Self) is directed at another character (orat the Dream Self), and the recipient somehowregisters it or reacts to it. Traditionally, in theHall & Van de Castle (1966) content analysissystem, social interactions have been classifiedinto three different categories: aggression,friendliness, and sexual interactions. It may be,however, that these three categories are toobroad, and do not cover or identify all theoretic-ally-interesting types of social interaction.

When it comes to the simulation of socialinteractions, the predictions of the SST should,again, be contrasted with the predictions de-rived from competing theories. In this case theSST needs to be distinguished from two othertheories: CH and the TST. The TST is a simu-lation theory that describes and explains thesimulation of aggressive behaviours in dreams,by including them under the category of“threatening events”. The function of dreaming,according to TST, is not to specialize in thesimulation of social interactions per se, but inthreatening events; thus, any social interactionsare simulated in dreams not because they aresocial events but because they are threateningevents. No independent social simulation theoryis required to explain the simulation of social in-teractions involving a threat; and aggressive be-haviours between dream characters are, obvi-ously, social interactions where the wellbeing ofthe Dream Self or some other dream characteris potentially threatened.

Compared to CH or SST, the TST can ac-count for the overrepresentation of threateningevents and aggressive interactions in dreams (ascompared to waking life, McNamara et al. 2005;Valli et al. 2008). The TST, however, gives nodescription or functional explanation for neutraland positive types of social interactions (unlessthey occur as parts of a threatening event). TheTST assumes that neutral and positive eventsin dreams are either parts of a threat simulation(e.g., responding to a threat by helping otherswho are targets of a threat) or that they repres-



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

ent some kind of superfluous, non-functionaldreaming that simply goes on automaticallyeven if the threat simulation mechanisms arenot activated. Thus, when it comes to social in-teractions, the SST should in particular predictand explain the neutral and friendly types of so-cial interactions, and show that some of themare actively selected as targets of dream simula-tion. In contrast, the CH predicts that neutraland positive types of social interactions shouldonly occur in the same proportions as they oc-cur in real life, passively reflecting their waking-life frequencies.

If, according to SST, the simulation ofneutral and positive social interactions indreams serve to represent and strengthen im-portant social connections and to rehearseprosocial behaviours in relation to those connec-tions, then these types of interactions shouldfrequently occur in dreams. This would servethe function of maintaining, rehearsing, orstrengthening our waking life social bonds andnetworks, and would satisfy our social need tobelong to groups that enhance our survival.After dreaming about prosocial behaviours, oursocial bonds during wakefulness would automat-ically be experienced as stronger and we wouldbe more likely to engage in behaviours that fur-ther strengthen those bonds. Some tentativesteps towards examining how the affects andcontents of social dreams predict subsequentwaking behaviour have been taken by Seltermanet al. (2014). They discovered that an increasedfrequency of dreams involving significant otherswas associated with higher levels of intimacyand interaction the following day, whereasdream infidelity predicted less intimacy. Repor-ted arguments in dreams were also found to becorrelated with subsequent conflict in wakinglife. They leave open the question whether thisis due to the conscious reflection of the report-ing procedure, a more implicit association, or amixture of the two.

Again, there are no detailed content ana-lysis studies that have investigated the exactnature of social interaction in dreams by takinginto account the social context of the interac-tion; that is, by studying who is engaged inwhat type of interaction and with whom. From

previous studies based on home dream diarieswe know that dreamer-involved aggression, ad-justed to take into account all social interac-tions except sexual interactions, is present in60% of male dreams and half (51%) of femaledreams (Domhoff 1996). When male strangersappear in a dream, the likelihood that physicalaggression will occur in that dream far exceedswhat would be expected on the basis of chance.Basically this means that male strangers signalphysical aggression. The dreamer, however, isan aggressor in 40% of male dreams and a thirdof all female dreams (Domhoff 1996).

Yet, as the Hall and Van de Castle normsindicate, there are friendly interactions indreams—slightly more often in female (42%)than male (38%) dreams (Domhoff 1996). Fe-males also dream more often of familiar people(58%) than of strangers (42%) while the oppos-ite is true for males (45% vs 55%, respectively);which might suggest that when there are morefamiliar people in dreams, there is also morefriendliness. The dreamer participates in themajority of interactions that involve friendliness(84% for females, 90% for males), and the be-friender proportion is 50% for males and 47%for females. Thus, both sexes initiate friendlyinteractions in their dreams approximately asoften as they are befriended. Helping and pro-tecting is the most frequent type of friendly be-haviour in both sexes, followed by friendly re-marks and compliments, and giving gifts orgranting loans. Surprisingly, however, there isvery little mutual or reciprocal friendliness, soalthough friendly interactions are initiated indreams by the Dream Self or other characters,in less than 10% of friendly interactions the actis reciprocated immediately. This observationgoes against any social simulation theory thatpredicts reciprocal friendliness should be highlyrepresented in dreams: this does not seem to bethe case.

McNamara et al. (2005) investigatedwhether types of social interaction are differentin REM than in NREM dreams compared towakefulness, and noticed that aggressive inter-actions were more often simulated in REMdreams, whereas friendly interactions were moreoften simulated in NREM dreams. Furthermore,



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

dreamer initiated friendliness was more typicalfor NREM than REM dreams. What is most in-teresting in this study, however, is that theyalso found that social interactions in general aremore often depicted in both REM and NREMdreams than in wake reports. While aggressionwas more often simulated in dreams than en-countered in waking life, the number of reportswith at least one occurrence of friendliness didnot differ significantly across sleep–wake states.Thus, these observations imply that dreams donot seem to overrepresent friendly interactionsas compared to waking experiences.

In sum, aggressive interactions seem to bemore prominent in dreams than neutral orfriendly interactions, which would lend moresupport to the TST than to SST, and friendlyinteractions are not more prominent in dreamsthan in waking life, which would lend supportto CH and the TST. Nevertheless, if simulationsare biologically functional, and if these twotypes of simulation functions are not mutuallyexclusive, might there be enough room in thedream content for simulation of neutral andpositive interactions, in such a way that it couldhave contributed to the inclusive fitness of ourdreaming ancestors?

6.4 Some testable ideas derived from SST

Let us see how this general approach to socialsimulation in dreams could be translated intosome directly testable hypotheses. Now, a gen-eral thesis derived from the SST could be for-mulated as follows:

Dreams are specialized in simulating themost important social connections and networksof the dreamer to give an additional selectiveadvantage and to enhance the survival of thedreamer in waking life. The simulations of par-ticular people (the frequency of their presencein a person’s dream life), and the simulations ofpositive interactions with particular people,should focus on the people closest to us in wak-ing life and on the social bonds most importantfor our inclusive fitness in the real world.

This thesis could be directly tested by de-riving some empirical predictions from it, telling

us what kind of simulations of social interac-tions and to what extent they should appear indreams. If dreams are specialized in the waypredicted by SST, then the most important so-cial networks and the people in them should ap-pear more frequently in dream life than in acorresponding stretch of waking life. That is,their frequency of occurrence should be targetsof active selection and inclusion into dreams,and hence over-represented and exaggerated indreams.

This empirical prediction could be testedby identifying a person’s most important socialnetworks in waking life, and by quantifying thefrequency of interactions of the dreamer withthose people during dreaming vs. during wake-fulness. In the already existing literature, thereare some data relevant to the hypothesis, butdata that directly compares waking social lifeand dream life in the manner required to testthe hypothesis seems to be lacking.

The data scattered in the literature de-scribes the relative frequency of dreams inwhich a certain type of close person appears onaverage in the dreams of the general (or the stu-dent) population. For example, romantic part-ners occur in 20% of dreams and this frequencycorrelates with the time spent together in wake-fulness (Schredl 2011; Schredl & Hofmann2003). Core family members occur in 10%–30%of dreams; parents in about 8%–20% of dreams,and siblings from 2%–7.5% of dreams (seeSchredl 2013). Friends occur in about 20% ofdreams (Roll & Millen 1979), but during long-term isolation from social contacts with friendsin one case (Merei 1994) this declined to 10%.In studies of long dream series from a singleperson, a close family member or spouse hasbeen found to be the person most oftendreamed about. In a sample of over two hun-dred dream reports, reported by a married wo-man (Arlie) with four grown-up children, themost frequently occurring character is her hus-band; whereas in a sample of over three hun-dred dreams from an unmarried woman in herthirties (Merri), the most frequently occurringcharacter is her sister, who was no longer aliveat the time when the dream reports were collec-ted (Schweickert 2007).



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

In Schredl’s studies, interesting analyses ofa long dream series from a single dreamer wereconducted, revealing the proportions of school-mates (2012) and family members (2013) simu-lated in dreams across a period stretching overtwenty years. Old school mates continued to ap-pear in about 5% of dreams over the years whenthe dreamer had nothing to do with them anymore in real life. Similarly, family members,even when the participant was not living withthem anymore, still retained a strong if some-what reduced presence in the same dreamseries, being present in approximately 15–20%of the dreams over a twenty-year period.

These results show that the probability ofoccurrence of a character in dreams is to someextent related to the amount of real life contactwith that person and to the closeness of the re-lationship in real life, thus supporting the CH.However, people who have at some point in lifebeen close and important do not seem to disap-pear totally from the dream simulations eventhough they have long ago totally disappearedfrom the real life of the dreamer. This feature ofthe already-existing data suggests that simula-tions of social contact might serve the functionof maintaining or strengthening close relation-ships over time. When the frequency of a previ-ously close and important social contact falls tozero in waking life, and the person is no longerencountered in waking life (like old school matesafter leaving school, or after the death of a fam-ily member), the simulation of such a personseems never to totally disappear from dreamlife, even if the frequency of dream simulationsof that person to some extent diminishes. Socialsimulations in dreams thus seem to maintain anactive storage and rehearsal of the most import-ant and closest social relationships of our entirelives, even when those relationships are brokenor discontinued for good, or are temporarily onhold in our waking lives.

What happens if a relationship that hasdisappeared from waking life is reactivated afteryears of disconnection? In Schredl’s (2012)study, old schoolmates met for a reunion twentyyears after going their separate ways. Interest-ingly, when the same relationships are re-activ-ated in real life for just one day, the dream sim-

ulation of those social relationships is increasedsignificantly and for a long period of time (com-pared to the time of actually meeting). Themechanism that reactivates old targets of simu-lation might be analogous to that proposed inTST for the re-activation of old threats. Thefrequency with which the most important realthreats are simulated (e.g., in post-traumaticnightmares) increases when, during wakefulness,new cues are encountered that are associatedwith the old threat possibly reoccurring in reallife.

These considerations suggest a more pre-cise function of social dream simulations thatcould be formulated along the following lines.We may call it the Strengthening Hypothesis:the function of social simulations in dreams isto maintain and strengthen the dreamer’s mostimportant social bonds from waking life. Con-sequently, a prediction derived from theStrengthening Hypothesis can be formulated asfollows: if strengthening important social bondsis a function of social dream simulations, thendreaming should include with high frequencysocial interactions in which the (current or past)most important social bonds are strengthenedthrough various types of simulated positive so-cial interactions and prosocial behaviours. Thus,the frequency of prosocial, positive interactions(bond-strengthening) with the most importantpersons should clearly surpass the frequency ofnegative (bond-weakening) interactions withindreams, and also be more frequent in dreamsthan in a corresponding stretch of waking life.

Schredl’s (2012, 2013) findings are to someextent consistent with both the CH and theSST, but do not allow any firm conclusionsabout which theory better predicts the occur-rence of the most important social connectionsin dreams. Studies that collect data from bothwaking life and dream life during the sameperiod of life from the same people, as well asfrom the life history of these individuals, are ne-cessarily required to test whether the represent-ation of the most important connections is ex-aggerated in dreams, or if they just reflect thewaking frequency. In practice, this predictioncould be tested by identifying all the interac-tions between the dreamer and the people in his



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

or her most important social networks, in bothdream and waking reports. Then the interac-tions could be classified according to whetherthey tend to strengthen or weaken the relation-ship with that particular person. If the fre-quency with which dreaming simulates positiveinteractions surpasses the frequency of those in-teractions in real life, then the SST would gaincredence over the CH.

Another potential simulation function toconsider can be called the Practise and Prepara-tion Hypothesis. According to this hypothesis,the function of social simulations in dreams isto force the dreamer to practise important so-cial bonding skills, such as how to give socialsupport to others. The prediction derived fromthis hypothesis states that if practising socialbonding skills is a function of dreaming, thenthe dreamer should frequently offer varioustypes of social support to other dream charac-ters, for example emotional, instrumental, or in-formational support. Furthermore, the types ofsocial support offered should be dependent onthe degree of relationship intimacy, i.e., the dis-tance between the self and the recipient in thehierarchy of the social world of the individual. Ifthe Practise and Preparation Hypothesis is cor-rect, then the frequency of simulating socialsupport should be higher than comparable be-haviours in real life.

These ideas are testable, but dream con-tent studies are to be carefully designed withthe specific aim of testing them. In the literat-ure already published, friendliness percentagesin different dream samples and descriptive stat-istics concerning who initiates friendliness indreams might shed some light on these ques-tions. However, without any data about the fre-quency of occurrence of these same behavioursin the waking state of the same person, thepurely descriptive findings from dreaming alonewill not be able to separate CH predictionsfrom SST predictions. The comparable wakingdata is crucial as a baseline against which thedream data can be evaluated and in relation towhich the CH predictions can be contrastedwith the SST predictions.

In an ideal setting the hypotheses for theSST and its proposed functions would also be

tested cross-culturally and in particular, as thetheory makes bold evolutionary claims, in tradi-tional small-scale human societies. As Henrichet al. (2010) have pointed out, the concentra-tion of behavioural research into the so-calledWestern, Educated, Industrialized, Rich, andDemocratic (WEIRD) societies are highly un-representative of the species, and might poseproblems for the generalizability of the results.Furthermore, by contrasting, for example, thedifferences between the social simulations ofsmall-scale and Western societies, we might un-cover useful information about the plasticityand ontogenetic mechanisms of the social simu-lation function.

7 Conclusions

The concept of “simulation” is a useful theoret-ical concept for dream research. It unifies defini-tions and descriptions of the basic nature ofdreaming, and helps to formulate testable theor-ies of the function of dreaming. Applying thisconcept to the social reality of dreams meansthat we start to describe the persons and socialinteractions in dreams as simulations of theircounterparts in real life. Consequently, we canask: How does the simulated social reality relateto the actual social reality in the same person’swaking life? Is it plausible to hypothesize thatthe avatars in the dreaming brain might in factbe there in order to force us to maintain andpractise various evolutionarily important func-tions of social perception and social bonding?

In this paper we made an attempt to cla-rify what it means to put forward the theoret-ical statement that “dreaming is a social simula-tion”, especially when this claim is offered as anexpression of a theory of the function of dream-ing. The SST can be formulated in a testablemanner, and a number of testable predictionscan be derived from it. Some of those predic-tions, concerning basic social perception andmindreading abilities, already receive ratherstrong support from the published literature.Many more hypotheses remain to be tested. Toachieve theoretically-informative results and todirectly contrast the predictions of different the-ories, future studies have to be designed in a



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

strictly theory-driven and hypothesis drivenmanner—which, unfortunately, is not a commonapproach in dream research.

If the SST, or some parts of it, prove suc-cessful, we have to be able to show that theSST predicts the nature and the occurrence ofsocial simulations in dreams more accuratelythan its main competitors, the CH and theTST. To fare better than the CH, the datawould have to show that the most important so-cial contents are actively selected for incorpora-tion in dreams as social simulations, and there-fore rehearsed in an exaggerated quantity orform in dreams. To show that the CH is on theright track, the data would have to show thatdream simulations merely reflect, both quantit-atively and qualitatively, whatever experienceswaking life has recently presented to the sameperson. To go beyond what the TST predictsand explains, the data supporting the SSTwould have to show that dreaming over-repres-ents and actively runs positive or neutral socialsimulations in dreams that strengthen the skillsof social perception and bonding, but that havenothing specifically to do with threat-perceptionand avoidance.

At this point, we are not yet sure howstrong the empirical case for SST is going to be,and whether the evidence will mostly turn outto be for or against it. We shall wait for thekind of studies that directly test SST and set itagainst other theories’ predictions. However,what we are confident about is that SST is anempirically testable theory, and that dream re-search would in general gain much if dream con-tent studies were rigorously designed to test thepredictions derived from opposing theories, andif dream data were in general collected and ana-lysed in a manner that provides us with strongtests of different theoretical hypotheses ratherthan just producing more and more purely de-scriptive data of dream content (and thenpresenting vague, post-hoc theoretical interpret-ations of them). In that way, dream researchwould be able to find and test new, promisingtheoretical ideas, perhaps derived from cognitiveand social neuroscience and from evolutionarypsychological considerations. New theoretically-guided studies would help leave behind old

ideas if they did not generate any clear andtestable predictions or if such predictions didnot gain sufficient empirical support.

Even if we will at some point be able toexplain some of the functions of social simula-tion in our dreams, we might not be able to ex-plain the underlying mechanisms that generatethe simulations. The fundamental metaphysicalnature of the simulated persons inhabiting ourdreaming brain might after all be almostequally mysterious as the immaterial nature ofa Cartesian ghost, because, like everything weexperience in our dreams, the avatars in ourdreams are built out of features that have noobjective, physically observable, or measurablesubstance. Instead, they consist of subjectively-experienced phenomenal features, and at leastat the present state of consciousness science, theonly way for us to get any empirically-baseddata about them is through the introspectivereports carefully collected from the dreamers.How the sleeping brain produces vivid, dy-namic, complex phenomenality and organizes itinto subjective spatiotemporal hallucinations,inhabited by avatars and social simulations, stillremains beyond any current theoretical explana-tions of dreaming and consciousness. Any plaus-ible explanation of the actual brain mechanismsthat do the trick would have to solve the hardproblem of consciousness (Chalmers 1996) andcross the explanatory gap (Levine 1983)between the objective neural mechanisms in thebrain and the subjective experiential realitiesgoing on in subjective consciousness. We are notquite there yet.

Acknowledgements

This research was supported by the Academy ofFinland, Research program HUMAN MIND,project number 266434.



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

References

Arnulf, I., Grosliere, L., Le Corvec, T., Golmard, J.-L.,Lascois, O. & Duguet, A. (2014). Will students pass acompetitive exam that they failed in their dreams?Consciousness & Cognition, 29, 36-47. 10.1016/j.concog.2014.06.010

Baumeister, R. F. & Leary, M. R. (1995). The need tobelong: Desire for interpersonal attachments as a fun-damental human motivation. Psychological Bulletin,117 (3), 497-529. 10.1037/0033-2909.117.3.497

Brereton, D. (2000). Dreaming, adaptation, and con-sciousness: The Social Mapping Hypothesis. Ethos, 28(3), 379-409. 10.1525/eth.2000.28.3.379

Bulkeley, K., Broughton, B., Sanchez, A. & Stiller, J.(2005). Earliest remembered dreams. Dreaming, 15 (3),205-222. 10.1037/1053-0797.15.3.205

Bulkeley, K. & Kahan, T. L. (2008). The impact ofSeptember 11 on dreaming. Consciousness and Cogni-tion, 17 (4), 1248-1256. 10.1016/j.concog.2008.07.001

Burnstein, E., Crandall, C. & Kitayama, S. (1994). Someneo-Darwinian decision rules for altruism: Weighingcues for inclusive fitness as a function of the biologicalimportance of the decision. Journal of Personality andSocial Psychology, 67 (5), 773-789. 10.1037/0022-3514.67.5.773

Chalmers, D. J. (1996). The conscious mind. Oxford: Ox-ford University Press.

Daly, M. & Wilson, M. (1988). Homicide. Hawthorne,NY: Aldine de Gruyter.

Daoust, A. M., Lusignan, F. A., Braun, C. M., Mottron,L. & Godbout, R. (2008). Dream content analysis inpersons with an autism-spectrum disorder. Journal ofAutism and Develpomental Disorders, 38 (4), 634-643.10.1007/s10803-007-0431-z

Deliens, G., Gilson, M. & Peigneux, P. (2014). Sleep andthe processing of emotions. Experimental Brain Re-search, 232 (5), 1403-1414. 10.1007/s00221-014-3832-1

Diamond, J. (2012). The world until yesterday. London,UK: Penguin.

Domhoff, G. W. (1996). Finding meaning in dreams: Aquantitative approach. New York, NY: Plenum.

(2007). Realistic simulation and bizarreness in dreamcontent: Past findings and suggestions for future research.In D. Barrett & P. McNamara (Eds.) The New Scienceof Dreaming (pp. 1-27). Westport, CT: Praeger.

Dunbar, R. I. M. (1992). Neocortex size as a constrainton group size in primates. Journal of Human Evolu-tion, 22 (6), 469-493. 10.1016/0047-2484(92)90081-J

(2008). Why humans aren’t just great apes. Issuesin Ethnology and Anthropology, 3 (3), 15-33.

Fonagy, P. & Target, M. (1997). Attachment and reflect-ive function: Their role in self-organization. Develop-ment and Psychopathology, 9 (4), 679-700.

Foulkes, D. (1985). Dreaming: A cognitive-psychologicalanalysis. Hillsdale, NJ: Lawrence Erlbaum.

Frank, R. H., Gilovich, T. & Regan, D. T. (1993). Theevolution of one-shot cooperation: An experiment. Eth-ology and Sociobiology, 14 (4), 247-256. 10.1016/0162-2095(93)90020-I

Franklin, M. S. & Zyphur, M. J. (2005). The role ofdreams in the evolution of the human mind. Evolution-ary Psychology, 3, 59-78.

Freud, S. (1950). The interpretation of dreams. NewYork, NY: Random House.

Gardner, W. L., Pickett, C. L. & Brewer, M. B. (2000).Social exclusion and selective memory: How the needto belong influences memory of social events. Personal-ity and Social Psychology Bulletin, 26 (4), 486-496.10.1177/0146167200266007

Germain, A., Shear, K. M., Walsh, C., Buysse, D. J.,Monk, T. H., Reynolds, C. F., Frank, E. & Silowash,R. (2013). Dream content in complicated grief: A win-dow into loss-related cognitive schemas. Death Studies,37 (3), 269-284. 10.1080/07481187.2011.641138

Hall, C. S. & Van de Castle, R. L. (1966). The contentanalysis of dreams. New York, NY: Appleton-Century-Crofts.

Hamilton, W. (1964). The genetical evolution of socialbehaviour. Journal of Theoretical Biology, 7 (1), 1-16.10.1016/0022-5193(64)90038-4

Hartmann, E. (1995). Making connections in a safe place:Is dreaming psychotherapy? Dreaming, 5 (4), 213-228.10-1037/h0094437

(1996). Outline for a theory on the nature andfunctions of dreaming. Dreaming, 6 (2), 147-170.10.1037/h0094452

(1998). Dreams and nightmares: The new theoryon the origin and meaning of dreams. New York, NY:Plenum Press.

Henrich, J., Heine, S. J. & Norenzayan, A. (2010). Theweirdest people in the world? Behavioral and BrainSciences, 33, 61-83. 10.1017/S0140525X0999152X

Heynick, F. (1993). Language and its disturbances indreams. New York, NY: Wiley.

Hill, K. R., Walker, R. S., Božičević, M., Eder, J., Head-land, T., Hewlett, B., Hurtado, M. & Wood, B. (2011).Co-residence patterns in hunter-gatherer societies show


http://dx.doi.org/10.1016/j.concog.2014.06.010

http://dx.doi.org/10.1037/0033-2909.117.3.497

http://dx.doi.org/10.1525/eth.2000.28.3.379

http://dx.doi.org/10.1037/1053-0797.15.3.205


http://dx.doi.org/10.1037/0022-3514.67.5.773

http://dx.doi.org/10.1007/s10803-007-0431-z

http://dx.doi.org/10.1007/s00221-014-3832-1

http://dx.doi.org/10.1016/0047-2484(92)90081-J

http://dx.doi.org/10.1016/0162-2095(93)90020-I

http://dx.doi.org/10.1177/0146167200266007

http://dx.doi.org/10.1080/07481187.2011.641138

http://dx.doi.org/10.1016/0022-5193(64)90038-4

http://dx.doi.org/10-1037/h0094437

http://dx.doi.org/10.1037/h0094452

http://dx.doi.org/10.1017/S0140525X0999152X


http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

unique human social structure. Science, 331 (6022),1286-1289. 10.1126/science.1199071

Hobson, J. A. (1988). The dreaming brain. New York,NY: Basic Books.

(1997). Dreaming as delirium: A mental statusexam of our nightly madness. Seminars in Neurology,17 (2), 121-128. 10.1055/s-2008-1040921

(2001). The dream drugstore: Chemically alteredstates of consciousness. Cambridge, MA: MIT Press.

(2009). REM sleep and dreaming: Towards a the-ory of protoconsciousness. Nature Reviews Neuros-cience, 10 (11), 803-813. 10.1038/nrn2716

(2011). Dream life. Cambridge, MA: MIT Press. Hobson, J. A., Pace-Schott, E. F. & Stickgold, R. (2000).

Dream science 2000: A response to commentaries onDreaming and the brain. Behavioral and Brain Sci-ences, 23 (6), 1019-1035. 10.1017/S0140525X00954025

Hobson, J. A. & Friston, K. J. (2012). Waking anddreaming consciousness: Neurobiological and functionalconsiderations. Progress in Neurobiology, 98 (1), 82-98.10.1016/j.pneurobio.2012.05.003

Hobson, J. A. & Schredl, M. (2011). The continuity anddiscontinuity between waking and dreaming: A dia-logue between Michael Schredl and Allan Hobson con-cerning the adequacy and completeness of these no-tions. International Journal of Dream Research, 4 (1),3-7. 10.11588/ijodr.2011.1.9087

Humphrey, N. (2000). Dreaming as play. Behavioral andBrain Sciences, 23 (6), 953-953.10.1017/S0140525X0054026

Kahn, D., Pace-Schott, E. & Hobson, J. A. (2002). Emo-tion and cognition: Feeling and character identificationin dreaming. Consciousness and Cognition, 11 (1), 34-50. 10.1006/ccog.2001.0537

Kahn, D. & Hobson, J. A. (2005). Theory of mind indreaming: Awareness of feelings and thoughts of othersin dreams. Dreaming, 15 (1), 48-57. 10.1037/1541-1559.15.1.48

Lara-Carrasco, J., Simard, V., Saint-Onge, K., Lam-oureux-Tremblay, V. & Nielsen, T. A. (2013). Maternalrepresentations in the dreams of pregnant women: aprospective comparative study. Frontiers in Psychology,4, 1-13. 10.3389/fpsyg.2013.00551

(2014). Disturbed dreaming during the third tri-mester of pregnancy. Sleep Medicine, 15 (6), 694-700.10.1016/j.sleep.2014.01.026

Leary, M. R., Tambor, E. S., Terdal, S. K. & Downs, D.L. (1995). Self-esteem as an interpersonal monitor: Thesociometer hypothesis. Journal of Personality and So-

cial Psychology, 68 (3), 518-530. 10.1037/0022-3514.68.3.518

Leary, M. R., Haupt, A. L., Strausser, K. S. & Chokel, J.T. (1998). Calibrating the sociometer: The relationshipbetween interpersonal appraisals and state self-esteem.Journal of Personality and Social Psychology, 74 (5),1290-1299. 10.1037/0022-3514.74.5.1290

Levine, J. (1983). Materialism and qualia: The explanat-ory gap. Pacific Philosophical Quarterly, 64, 354-361.

Llinás, R. (2001). I of the vortex: From neurons to self.Cambridge, MA: MIT Press.

Llinás, R. R. & Paré, D. (1991). Of dreaming and wake-fulness. Neuroscience, 44 (3), 521-535. 10.1016/0306-4522(91)90075-Y

Llinás, R. & Ribary, U. (1994). Perception as an oneiric-like state modulated by the senses. In C. Koch & J. L.Davis (Eds.) Large-scale neuronal theories of the brain(pp. 111-124). Cambridge, MA: MIT Press.

McNamara, P. (1996). REM sleep: A social bondingmechanism. New Ideas in Psychology, 14 (1), 35-46.10.1016/0732-118X(95)00023-A

(1996). REM sleep: A social bonding mechanism.New Ideas in Psychology, 14 (1), 35-46. 10.1016/0732-118X(95)00023-A

McNamara, P., McLaren, D., Smith, D., Brown, A. &Stickgold, R. (2005). A “Jekyll and Hyde” within: ag-gressive versus friendly interactions in REM and non-REM dreams. Psychological Science, 16 (2), 130-136.10.1111/j.0956-7976.2005.00793.x

McNamara, P., McLaren, D., Kowalczyk, S. & Pace-Schott, E. (2007). “Theory of mind” in REM andNREM dreams. In D. Barrett & P. McNamara (Eds.)The new science of dreaming (pp. 201-220). Westport,CT: Praeger.

Mealey, L., Daood, C. & Krage, M. (1996). Enhancedmemory for faces of cheaters. Ethology and Sociobio-logy, 17 (2), 119-128. 10.1016/0162-3095(95)00131-X

Merei, F. (1994). Social relationships in manifest dreamcontent. Journal of Russian and East European Psy-chology, 32 (1), 46-88. 10.2753/RPO1061-0405320146

Metzinger, T. (2003). Being no one: The self -model the-ory of subjectivity. Cambridge, MA: MIT Press.

Nielsen, T. A. (2000). A review of mentation in REM andNREM sleep: “Covert” REM sleep as a possible recon-ciliation of two opposing models. Behavioral and BrainSciences, 23 (6), 851-866. 10.1017/S0140525X0000399X

(2010). Dream analysis and classification: Thereality simulation perspective. In M. Kryeger, T. Roth& W. C. Dement (Eds.) Principles and Practice of


http://dx.doi.org/10.1126/science.1199071

http://dx.doi.org/10.1055/s-2008-1040921

http://dx.doi.org/10.1038/nrn2716

http://dx.doi.org/10.1017/S0140525X00954025

http://dx.doi.org/10.1016/j.pneurobio.2012.05.003

http://dx.doi.org/10.11588/ijodr.2011.1.9087

http://dx.doi.org/10.1017/S0140525X0054026

http://dx.doi.org/10.1006/ccog.2001.0537

http://dx.doi.org/10.1037/1541-1559.15.1.48

http://dx.doi.org/10.3389/fpsyg.2013.00551

http://dx.doi.org/10.1016/j.sleep.2014.01.026

http://dx.doi.org/10.1037/0022-3514.68.3.518

http://dx.doi.org/10.1037/0022-3514.74.5.1290

http://dx.doi.org/10.1016/0306-4522(91)90075-Y

http://dx.doi.org/10.1016/0732-118X(95)00023-A

http://dx.doi.org/10.1016/0732-118X(95)00023-A

http://dx.doi.org/10.1016/0732-118X(95)00023-A

http://dx.doi.org/10.1111/j.0956-7976.2005.00793.x

http://dx.doi.org/10.1016/0162-3095(95)00131-X

http://dx.doi.org/10.2753/RPO1061-0405320146



http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

Sleep Medicine (pp. 595-603). New York, NY: Elsevier. Nielsen, T. A. & Germain, A. (2000). Post-traumatic

nightmares as a dysfunctional state. Behavioral andBrain Sciences, 23 (6), 978-979.10.1017/S0140525X0070402X

Nielsen, T. A. & Lara-Carrasco, J. (2007). Nightmares,dreaming, and emotion regulation. In D. Barrett & P.McNamara (Eds.) The new science of dreaming (pp.253-284). Westport, CT: Praeger.

Nielsen, T. A & Levin, R. (2007). Nightmares: A newneurocognitive model. Sleep Medicine Reviews, 11 (4),295-310. 10.1016/j.smrv.2007.03.004

Nielsen, T. A. & Powell, R. A. (1989). The “dream-lag”effect: A 6-day temporal delay in dream content incor-poration. Psychiatric Journal of the University of Ott-awa, 14 (4), 561-565.

Oberman, L. M. & Ramachandran, V. S. (2007). Thesimulating social mind: The role of the mirror neuronsystem and simulation in the social and communicativedeficits of autism spectrum disorders. PsychologicalBulletin, 133 (2), 310-327. 10.1037/0033-2909.133.2.310

Paul, F. & Schredl, M. (2012). Male-female ratio in wak-ing-life contacts and dream characters. InternationalJournal of Dream Research, 5 (2), 119-124.10.11588/ijodr.2012.2.9406

Pickett, C. L., Gardner, W. L. & Knowles, M. (2004).Getting a cue: The need to belong and enhanced sens-itivity to social cues. Personality and Social PsychologyBulletin, 30 (9), 1095-1107. 10.1177/0146167203262085

Rechtschaffen, A. & Buchignani, C. (1992). The visualappearance of dreams. In J. S. Antrobus & M. Bertini(Eds.) The neuropsychology of sleep and dreaming (pp.143-155). Hillsdale, NJ: Lawrence Erlbaum.

Revonsuo, A. (1995). Consciousness, dreams, and virtualrealities. Philosophical Psychology, 8 (1), 35-58.10.1080/095115089508573144

(2000). The reinterpretation of dreams: An evolu-tionary hypothesis of the function of dreaming. Behavi-oral and Brain Sciences, 23 (6), 877-901.10.1017/S0140525X00004015

(2005). The self in dreams. In T. E. Feinberg & J.P. Keenan (Eds.) The lost self: Pathologies of the brainand mind (pp. 206-219). New York, NY: Oxford Uni-versity Press.

(2006). Inner presence: Consciousness as a biolo-gical phenomenon. Cambridge, MA: MIT Press.

Revonsuo, A. & Salmivalli, C. (1995). A content analysisof bizarre elements in dreams. Dreaming, 5 (3), 169-187. 10.1037/h0094433

Revonsuo, A. & Tarkko, K. (2002). Binding in dreams.Journal of Consciousness Studies, 9 (7), 3-24.

Robert, G. & Zadra, A. (2014). Thematic and contentanalysis of idiopathic nightmares and bad dreams.Sleep, 37 (2), 409-417. 10.5665/sleep.3426

Roll, S. & Millen, L. (1979). The friend as represented inthe dreams of late adolescents: Friendship withoutrose-coloured glasses. Adolescence, 14 (54), 255-275.

Schredl, M. (2011). Dreams of a romantic partner in adream series: Comparing relationship periods withperiods of being separated. International Journal ofDream Research, 4 (2), 127-131.10.11588/ijodr.2011.2.9150

(2012). Old school friends: Former social relation-ship patterns in a long dream series. InternationalJournal of Dream Research, 5 (2), 143-147.10.11588/ijodr.2012.2.9432

(2013). Dreams of core family members in a longdream series. International Journal of Dream Research,6 (2), 114-118. 10.11588/ijodr.2013.2.11055

Schredl, M. & Hofmann, F. (2003). Continuity betweenwaking activities and dream activities. Consciousnessand Cognition, 12 (2), 298-308. 10.1016/S1053-8100(02)00072-7

Schweickert, R. (2007). Social networks of characters indreams. In D. Barrett & P. McNamara (Eds.) The newscience of dreaming. Westport, CT: Praeger.

Selterman, D. & Drigotas, S. (2009). Attachment stylesand emotional content, stress, and conflict in dreams ofromantic partners. Dreaming, 19 (3), 135-151.10.1037/a0017087

Selterman, D. F., Apetroaia, A. I., Riela, S. & Aron, A.(2014). Dreaming of you: Behavior and emotion indreams of significant others predict subsequent rela-tional behaviour. Social Psychological and PersonalityScience, 5 (1), 111-118. 10.1177/1948550613486678

Strauch, I. & Meier, B. (1996). In search of dreams: Res-ults of experimental dream research. New York, NY:SUNY Press.

Stumbrys, T., Erlacher, D. & Schmidt, S. (2011). Luciddream mathematics: An explorative online study ofarithmetic abilities of dream characters. InternationalJournal of Dream Research, 4 (1), 35-40.10.11588/ijodr.2011.1.9079

Sutcliffe, A., Dunbar, R., Binder, J. & Arrow, H. (2012).Relationships and the social brain: Integrating psycho-logical and evolutionary perspectives. British Journalof Psychology, 103 (2), 149-168. 10.111/j.2044-8295.2011.02061.x



http://dx.doi.org/10.1016/j.smrv.2007.03.004

http://dx.doi.org/10.1037/0033-2909.133.2.310


http://dx.doi.org/10.1177/0146167203262085

http://dx.doi.org/10.1080/095115089508573144

http://dx.doi.org/10.1017/S0140525X00004015

http://dx.doi.org/10.1037/h0094433

http://dx.doi.org/10.5665/sleep.3426




http://dx.doi.org/10.1016/S1053-8100(02)00072-7

http://dx.doi.org/10.1037/a0017087

http://dx.doi.org/10.1177/1948550613486678


http://dx.doi.org/10.111/j.2044-8295.2011.02061.x


http://dx.doi.org/10.15502/9783958570375


www.open-mind.net

Tholey, P. (1989). Consciousness and abilities of dreamcharacters observed during lucid dreaming. Perceptualand Motor Skills, 68 (2), 567-578.10.2466/pms.1989.68.2.567

Trivers, R. L. (1971). The evolution of reciprocal altru-ism. The Quarterly Review of Biology, 46 (1), 35-57.

Uguccioni, G., Golmard, J. L., de Fontréaux, A. N., Leu-Semenescu, S., Brion, A. & Arnulf, I. (2013). Fight orflight? Dream content during sleepwalking/sleep terrorsvs. rapid eye movement sleep behavior disorder. SleepMedicine, 14 (5), 391-398. 10.1016/j.sleep.2013.01.014

Valli, K., Revonsuo, A., Pälkäs, O. & Punamäki, R.-L.(2006). The effect of trauma on dream content: A fieldstudy of Palestinian children. Dreaming, 16 (2), 63-87.10.1037/1053-0797.16.2.63

Valli, K., Strandholm, T., Sillanmäki, L. & Revonsuo, A.(2008). Dreams are more negative than real life: Im-plications for the function of dreaming. Cognition andEmotion, 22 (5), 833-861. 10.1080/02699930701541591

Valli, K. & Revonsuo, A. (2006). Recurrent dreams: Re-curring threat simulations? Consciousness and Cogni-tion, 15 (2), 470-474. 10.1016/concog.2005.05.2001

(2009). The threat simulation theory in light of re-cent empirical evidence: A review. American Journal ofPsychology, 122 (1), 17-38.

Windt, J. M. (2010). The immersive spatiotemporal hal-lucination model of dreaming. Phenomenology andCognitive Science, 9 (2), 295-316. 10.1007/s11097-010-9163-1

Windt, J. M. & Metzinger, T. (2007). The philosophy ofdreaming and self-consciousness: What happens to theexperiential subject during the dream state? In D. Bar-rett & P. McNamara (Eds.) The new science of dream-ing (pp. 193-247). Westport, CT: Praeger.

Zadra, A., Desjardins, S. & Marcotte, E. (2006). Evolu-tionary function of dreams: A test of the threat simula-tion theory in recurrent dreams. Consciousness andCognition, 15, 450-463. 10.1016/j.concog.2005.02.002


http://dx.doi.org/10.2466/pms.1989.68.2.567

http://dx.doi.org/10.1016/j.sleep.2013.01.014

http://dx.doi.org/10.1037/1053-0797.16.2.63

http://dx.doi.org/10.1080/02699930701541591

http://dx.doi.org/10.1016/concog.2005.05.2001

http://dx.doi.org/10.1007/s11097-010-9163-1



http://dx.doi.org/10.15502/9783958570375


the avatars in the machine

Documents