- Email: [email protected]
Dreaming in the multilevel framework
Consciousness and Cognition 20 (2011) 1084–1090
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Dreaming in the multilevel framework q Katja Valli ⇑ Centre for Cognitive Neuroscience, University of Turku, Finland School of Humanities and Informatics, University of Skövde, Sweden
a r t i c l e
i n f o
Article history: Available online 13 May 2011 Keywords: Biological realism Multilevel framework Dreaming Phenomenal consciousness
a b s t r a c t Biological realism (Revonsuo, 2001, 2006) states that dreaming is a biological phenomenon and therefore explainable in naturalistic terms, similar to the explanation of other biological phenomena. In the biological sciences, the structure of explanations can be described with the help of a framework called ‘multilevel explanation’. The multilevel model provides a context that assists to clarify what needs to be explained and how, and how to place different theories into the same model. Here, I will argue that the multilevel framework would be useful when we try to construct scientific explanations of dreaming. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction Dreaming is the most often naturally occurring altered states of consciousness, experienced on a nightly basis by most individuals (Pagel, 2003; Strauch & Meier, 1996; Wallach, 1963). Dreaming is thus a universal phenomenon, and the brain’s capability to produce subjective hallucinatory experiences during sleep seems to be a part of our biological machinery, a trait genetically passed on in successive generations. Nevertheless, despite rigorous empirical research and theorizing, it is still far from clear why our brain produces these experiences, and how they are created. 2. Biological realism and the multilevel framework Biological realism (Revonsuo, 2001, 2006) treats consciousness as a real natural biological phenomenon that ‘‘literally resides within the confines of the brain’’ (Revonsuo, 2006, p. 10).1 Thus, it follows that dreaming as a form of subjective consciousness can be given a fully naturalistic explanation, similar to the explanation of other biological phenomena. Consequently, the explanatory framework used in biological sciences could also be used in dream research as well as more generally in consciousness research.2 In biological sciences the underlying explanatory strategy is ‘multilevel explanation’ (e.g., Bechtel, 1994; Bechtel & Richardson, 1992, 1993; Craver, 2001; Craver & Darden, 2001; Tinbergen, 1963). Basically, biological systems are hierarchical, composed of multiple but interconnected levels of organization. The multilevel model is utilized to conceptualize a framework for explaining the phenomenon under investigation, and it functions as a context that helps to clarify what needs to be explained and how, and how to place different theories into the same model (see Fig. 1). q This article is part of a special issue of this journal on European Science Foundation EMRC Exploratory Workshop: The Dreaming Mind-Brain, Consciousness and Psychosis (Challand Saint Anselme, Italy, 25th – 28th May 2009). ⇑ Address: Centre for Cognitive Neuroscience, Department of Behavioural Sciences and Philosophy/Psychology, Assistentinkatu 7, FI-20014 University of Turku, Finland. Fax: +358 23338770. E-mail address: katval@utu.fi 1 An approach very close to biological realism is biological naturalism, proposed by Searle (1992). In philosophy, however, biological views to consciousness are in the minority, and there are a number of other contemporary approaches, with varying philosophical commitments (e.g., Chalmers, 1996; Dennett, 1991; Dretske, 1995; O’Regan & Noë, 2001; Thompson, 2001; Tye, 1995). 2 For more details on how to apply the multilevel model to consciousness research in general, see Revonsuo (2006).
1053-8100/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2011.04.004
K. Valli / Consciousness and Cognition 20 (2011) 1084–1090
1085
Fig. 1. Phenomenon in the multilevel framework. Modified from Revonsuo (2006) and Valli (2008).
In the multilevel framework, the phenomenon under investigation is first identified and described at one level of organization (Revonsuo, 2006). The aim is to give a detailed account of what the phenomenon of interest is like. The next step involves figuring out what the components constituting and contributing to the phenomenon are. The explanation of the phenomenon can be broken down into several smaller explanatory tasks; constitutive, etiological, and contextual explanations. Each of these addresses different level of explanation in the hierarchical system (Revonsuo, 2006). Constitutive (the downward-looking) explanations involve looking at smaller-scale phenomena or a lower level of organization. They aim at decomposing the phenomenon under investigation to its constituent parts, and describing the lower level mechanisms that causally contribute to the emergence of the phenomenon. Etiological (the backward-looking) explanations focus on the temporal and causal history of the phenomenon, aiming to trace backward in time the causal chains that led to the occurrence of the phenomenon. Contextual (the upward-looking) explanations, in contrast, focus on describing the role of the phenomenon at higher level of organization where it is just one element or subsystem in a complex mechanism. Each level of explanation in the hierarchical structure, forming now a subproblem in the overall explanation of the phenomenon, can then be separately investigated by a dedicated science or with a particular method (Revonsuo, 2006). When sufficient answers to research problems on all or most levels of the framework have been acquired, we can begin to build an idealized multilevel model of the phenomenon under investigation. The main function of this model is to reveal the structure and function of the phenomenon across the hierarchical levels. As in multilevel explanation the phenomenon is approached from various levels, crossing boundaries between academic disciplines, it can be the basis for the development of an interdisciplinary discussion and consensus. 3. Dreaming in the multilevel framework If we treat dream consciousness as residing on one level of organization in a biological system, then to fully understand the phenomenon, we should insert dreaming into a multilevel framework as described in Fig. 2. First, the phenomenon itself needs to be identified, defined, and described in detail. This task is the responsibility of datadriven, descriptive dream research. The methodology required to accomplish this task is the systematic sampling of dreaming from various populations, and detailed content analyses of dream data that reveal the statistical patterns of the form and content of dream experiences. We already know a great deal, but controversies remain. Even the definition of dreaming has proven to be problematic (e.g., Nielsen, 2000), and therefore the descriptions of what dreaming is like also vary. Thus, the first task of dream science is to come to a consensus on what dreaming is, and in the case this can be accomplished, formulate a description of the phenomenon that majority agree on. 3.1. The constitutive level The constitutive explanation involves decomposing dream experience into its constituent parts, and describing the lower level mechanisms that bring about dreaming or whose activity together produces the phenomenon. Thus, the downwardlooking explanations need to describe the underlying mental, cognitive, and neural nonconscious mechanisms, both
1086
K. Valli / Consciousness and Cognition 20 (2011) 1084–1090
Fig. 2. Dreaming in the multilevel framework. Modified from Valli (2008).
structural and functional, that are the substrates for the form and content of dreams. With constitutive explanations, we run first into the ‘easy problem’ of consciousness (Chalmers, 1996). The easy problem concerns the neural or cognitive correlates of consciousness, and is mainly empirical. The constitutive level explanations for dreaming involve the ‘dream generator system’ (see also Fig. 1). Currently, mainly the Random Activation Theories (RATs) (e.g. Flanagan, 1995, 2000; Hobson, 1988; Hobson & McCarley, 1977; Hobson, PaceSchott, & Stickgold, 2000) address the constitutive mechanisms of dream consciousness in detail. In brief, RATs state that the mechanisms that actively generate REM sleep state also participate in generating dreams. The REM sleep regulating mechanisms are located in the brainstem, and the sleep related activation (for example, the ponto-geniculo-occipital (PGO) – waves) leaks over to the emotional and memory networks resulting in the random activation of these systems. The activated memories and emotions are synthesized into dream images by the forebrain that attempts to make sense of them. The dream generator system, however, is probably distinct from the REM sleep generator system, as dreaming and sleep stages can occur independently of each other.3 Therefore, dreaming should not be approached by focusing merely on the physiological aspects of different sleep stages, but rather we should look for the same underlying mechanism that produces dreaming despite of differences in brain activation patterns during different phases of sleep. Thus far, the constitutive mechanisms for dreaming, as for waking consciousness, have not been found. We know neither how the brain creates dream experiences nor where this process takes place. The methods available in cognitive neuroscience, especially as technologies evolve, can contribute to the identification and localization of the dream generator mechanism in the brain, and add to the understanding of constitutive explanations of dreaming. Nevertheless, the constitutive explanations for dream consciousness should not be misidentified with the neural correlates of dream consciousness. The constitutive mechanisms must go beyond correlative relationships, and specify the underlying hierarchical components that bring about dream consciousness. With full description of the constitutive mechanisms, we might also be able to address the ‘hard problem’ of consciousness (Chalmers, 1996), that is, by explaining the performance of all the relevant neural and cognitive systems and functions underlying phenomenal consciousness, we would know why physical processes in the brain give rise to the inner presence of mind. 3.2. The etiological level An etiological explanation aims to clarify the temporal and causal history of the phenomenon, and thus, the etiological explanations can be divided into (1) the immediate or proximate causes of the phenomenon, (2) the ontogenetic or developmental causes of the phenomenon, and (3) the ultimate or evolutionary causes of the phenomenon. In sum, the
3 Or vice versa, the currently used sleep stage classification systems are too crude to pick up similarities between sleep stages that may underlie dreaming, see, e.g., the hypothesis of covert REM processes operating in NREM sleep by Nielsen (2000).
K. Valli / Consciousness and Cognition 20 (2011) 1084–1090
1087
etiological explanations track back in time the causal chain that leads to the phenomenon. In the context of dreaming, these explanations involve neural and psychological level causes, as well as evolutionary biological causes addressing the phylogeny and possible adaptive function of dreaming. 3.2.1. The proximate causes The proximate causes of dreaming in the neural level should describe the temporal activation of the dream generator system in the brain and how it produces the internal stimulation that leads to subjective dream experience. Psychological level proximate explanations should describe how the immediate cognitive level preceding causes, such as the activation of memory sources, emotional state and activity, hyperassociativity, memory consolidation, attention, stimulus incorporation, and self-awareness, participate in producing dream phenomena. The ontogenetic or developmental explanations in the context of dreaming should aim to clarify how, during individual development from an embryo to adulthood, the brain becomes wired up so that the dream generator system matures and how the mind develops cognitively and psychologically so that it gradually produces adult-like dream experiences. Finally, the ultimate explanations should reveal the phylogenetic history of dreaming, and address the question whether dreaming was selected for its adaptive function or whether it is a spandrel or a mere random effect. The majority of dream theories include descriptions of hypothesized proximate etiological level mechanisms, and some also address the developmental level. Nevertheless, the exact causal chain of dream generation is unclear. For example, RATs (Hobson, 1988; Hobson & McCarley, 1977; Hobson et al., 2000) propose that the activation of the sleep-related mechanisms triggers the random activation of the emotional and memory networks. As the most recent memory traces encoded into the long term memory are high in saliency, and thus primed in the network, the random activation of the networks leads to the activation of most recent memories. Therefore, day residues and recent events are more often represented in dreams than events long past. The randomly activated memories are then bound into dream narratives, as the forebrain is trying to interpret the internal brain activation as it normally interprets external stimulation while awake. The narrative and meaningful nature of dream experience follows, on one hand, from the attempts by the forebrain to produce a coherent experience from the chaotic inputs received from the brain stem which has led to random memory activation. On the other hand, the binding process fails often, leading to hyperassociativity and bizarre dream features. Foulkes’s (1985, 1999) Cognitive Theory acknowledges that internal brain activation during sleep is the primary source of memory activation. Thus, Foulkes’s model seems to be compatible with the RATs in the sense that dream content is constructed from randomly or semi-randomly activated memories. Similarly, as the Continuity hypothesis (CH) (Domhoff, 2003; Schredl, 2000; Schredl & Hofmann, 2003) states that dreaming reflects wake experiences, the waking experiences stored into long term memory must be activated during sleep. Whether this activation is random or organized, the CH does not specify (although the emotional concerns of the dreamer may to some degree affect the activation of memory traces), but nevertheless, it leads to an organized and coherent dream experience. CH does neither describe how or why exactly dream content is generated from the activation. The Psychological Theories (e.g., Cartwright, 1996; Hartmann, 1995, 1996, 1998; Kramer, 1991, 1993) assume that dream content is not randomly constructed, but is based on specific activation patterns of most salient memory traces. The Threat Simulation Theory (TST) (Revonsuo, 2000) also proposes that selective and organized neural and cognitive processes are required in order for dreaming to take place. Although the theories do not explicate how exactly dreams are constructed, specifically the emotional concerns of the dreamer, or the most negatively charged or most recently activated memory traces guide dream content construction. 3.2.2. The ontogenetic causes The ontogenetic explanations for dreaming should clarify how, during individual development, the brain and mind mechanisms develop and mature so that an individual begins to dream and eventually have adult-like dreams. The Cognitive Theory (Foulkes, 1982, 1985, 1999) and the TST (Revonsuo, 2000) address developmental level etiological explanations. Children’s dreaming, according to Foulkes’s (1982, 1985) laboratory studies with awakenings from NREM and REM sleep stages is different from adults’ dreaming, because it is dependent on memory and thought processes which develop with age. Thus, dreaming follows developmental trends. The developmental level explanations for dreaming put forth by the TST are that dreaming has a genetic basis, and that dreaming reliably matures over the developmental life history of the organism according to the specifications contained in its genetic code. The maturation rate and the activity level of the threat simulation system is highly dependent on the activity level of the dream production mechanisms and the level of ecologically valid threat cue input, that is, personal threat experiences. 3.2.3. The ultimate causes The ultimate or evolutionary explanations aim to elucidate how, at some point in the evolutionary history, the human brain evolved the ability to produce dreaming. When it comes to dreaming, evolutionary explanations try to figure out how the first dream generator appeared in some distant ancestors of ours, perhaps in the first humans, perhaps in the first mammals. The phylogeny of dreaming is currently unclear, and as subjective experiences based on introspective reports can only be investigated in humans, comparative studies in other mammalian species are impossible. Dreaming may, nevertheless, be present in other species. For example, the TST acknowledges that animals may have subjective experiences during sleep to the same extent that they have subjective experiences during wakefulness (Revonsuo, 2000).
1088
K. Valli / Consciousness and Cognition 20 (2011) 1084–1090
When evolutionary psychological reasoning is extended to dreaming, the following questions about the adaptive function of dreaming emerge: Is it conceivable that the process of dreaming or dreaming about particular contents was functional in the biological sense in the ancestral world - that somehow our ancestors (independent of whether dreaming emerged in human, hominid or even more distant ancestors) had better chances of surviving or reproducing because they regularly dreamed during the night, and especially because they regularly dreamed about particular things? Could dreaming have given any kind of selective advantage for our ancestors and if yes, what kind of advantage? Does dreaming fulfill the main criterion for an adaptation, in other words, does dreaming show a clear design for a specific function, and does it seem to be specifically engineered to solving a specific adaptive problem, or did it appear as a nonfunctional epiphenomenon of other sleep-related mechanisms, as a so-called spandrel? In the past ten years, several hypothetical answers to the above questions have been presented. Dreaming has been suggested to serve adaptive functions similar to mammalian play behaviors (Bulkeley, 2004; Cheyne, 2000; Humphrey, 2000; Peterson & DeYoung, 2000). In Brereton’s (2000) ‘Social mapping hypothesis’ dreaming is suggested to have rehearsed the perceptual and emotional features required by successful social mapping, for example, by simulating the self, location, and awareness of others including awareness of their internal mental states. This is proposed to have eventually led to the emergence of self-awareness. Relatedly, Kahn and Hobson (2005) emphasize that awareness of what others are thinking and feeling is a robust aspect of human consciousness, and this aspect is maintained during dreaming. Thus, even though they do not explicitly express it, they imply that awareness of the minds of others (‘‘Theory of Mind’’) during dreaming might have contributed to the ability to anticipate the intentions of others while awake. Humphrey (2000), and Nielsen and Germain (2000) have proposed that as many of the selection pressures faced by ancestral humans were posed by complex human social life, modeling human relationships and interpersonal bonds might have had adaptive value. Franklin and Zyphur’s (2005) argument is organized along the similar lines; they propose that via dreaming it is possible to practice dealing with complex social situations, because those most adept in their social environment were likely to have the best access to resources in their social group, such as mates. McNamara (2004, 2007) suggests that costly REM features affect dream content, and through producing hard to fake emotional signals, increase fitness by facilitating cooperative social interactions or courtship displays. The psychological adaptation and creativity functions assigned for dreaming (Hartmann, 1996, 1998) suggest, though in a subtle manner, that the emotional problem solving and creativity function of dreams might have fulfilled biological functions in the ancestral environment. The TST (Revonsuo, 2000), in contrast, states that dreaming about threatening situations led to enhanced threat recognition and avoidance, thus enhancing survival and offspring production. If dream consciousness has an adaptive function, then it follows that dreaming has, evolutionarily speaking, causal powers. If phenomenal level experiences promoted the (inclusive) fitness of the organism, they must have affected behavior and organism-environment interaction in such a way that fitness promoting behaviors were more likely or more effective. Behavior and organism-environment interactions, however, belong to the contextual level explanations. 3.3. The contextual level The contextual level explanations describe the role of the phenomenon at a higher level of organization. Does dreaming relate to or affect other psychological or conscious/nonconscious phenomena and how? What consequences, if any, does the dream generator system have for learning, memory, emotion regulation, perception, and ultimately, overt behavior? In the contextual level, mental phenomenona figure in the etiological explanations for observable behavior. Thus, contextual explanations tie dreaming to other subjective and psychological phenomena on a more general level, which are intra- or interindividual level, overt behavior, and organism-environment interaction. Further, in evolutionary explanations for dreaming, the contextual level explanations need to be tied to the ultimate etiological level explanations. Contextual level explanations are addressed in many dream theories. The theories assigning a biological function for dreaming all provide a contextual level explanation, as dreaming must have modified waking consciousness or behavior in some way. The specific proposed effects range from emergence of self-awareness (Brereton, 2000) to enhancement of social skills (Franklin & Zyphur, 2005; Humphrey, 2000; Nielsen & Germain, 2000) and to threat avoidance behaviors (Revonsuo, 2000). Recently, dream imagery has also been linked to more effective sleep dependent memory consolidation (Wamsley, Tucker, Payne, Benavised, & Stickgold, 2010). In addition, Foulkes (1982, 1985, 1999) suggested that visuo-spatial skills are utilized in dream construction, and correlate with the complexity of dream content. The Psychological Theories assume that the very reason for why we dream lies in the usefulness of dream imagery for the psychological well-being of the individual (Cartwright, 1996; Hartmann, 1995, 1996, 1998), or that dreaming has a mood regulatory function (Kramer, 1991, 1993). In these theories, general psychological well-being is at the contextual level. 4. Conclusions and future directions The stand point taken in this paper is that dreaming is a biological phenomenon residing in the brain, and it can thus be explained in a similar fashion as any other biological phenomenon. Therefore, the multilevel framework can be used as a metatheory, or a model, to compare and evaluate the scope, limits, and contradictions of the more specific primary theories of dreaming, and identify conflicts between, lack of explanation, and evidence for specific theories within the same
K. Valli / Consciousness and Cognition 20 (2011) 1084–1090
1089
explanatory dimension of the model. This kind of an analysis reveal that the majority of dream theories do not address all the levels of explanation explicated in the multilevel framework, and in addition, the offered explanations are sometimes contradictory. If we are ever to build an overarching explanation and agree on what dreaming is, we need to reach reconciliation between contradictory views. The multilevel framework seems to be a promising basis for dream research, inviting contributions from various academic fields, and enabling interdisciplinary approaches to the development of an overarching theory of dreaming. The main reason for using a grounded theoretical model in dream studies is that without a model, such as the multilevel framework, that guides researchers to come up with holistic explanations, we will likely end up like the blind men studying an elephant from different perspectives, coming to completely different conclusions as to what the elephant is.
Acknowledgments I wish to thank Prof. Antti Revonsuo for his valuable comments during the preparation of the manuscript. This work was supported by the Academy of Finland (project 2600016311).
References Bechtel, W. (1994). Levels of description and explanation in cognitive science. Minds and Machines, 4, 1–25. Bechtel, W., & Richardson, R. C. (1992). Emergent phenomena and complex systems. In A. Beckermann, H. Flohr, & J. Kim (Eds.), Emergence or reduction? Essays on the prospects of nonreductive physicalism (pp. 257–288). Berlin: de Gruyter. Bechtel, W., & Richardson, R. C. (1993). Discovering complexity. Decomposition and localization as strategies in scientific research. Princeton, NJ: Princeton University Press. Brereton, D. (2000). Dreaming, adaptation, and consciousness: The social mapping hypothesis. Ethos, 28, 379–409. Bulkeley, K. (2004). Dreaming is play II: Revonsuo’s threat simulation theory in ludic context. Sleep and Hypnosis, 6, 119–129. Cartwright, R. D. (1996). Dreams and adaptation to divorce. In D. Barrett (Ed.), Trauma and dreams (pp. 173–185). Cambridge, MA: Harvard University Press. Chalmers, D. J. (1996). The conscious mind. Oxford: Oxford University Press. Cheyne, J. A. (2000). Play, dreams, and simulation. Behavioral and Brain Sciences, 23(6), 918–919. Craver, C. F. (2001). Role functions, mechanisms, and hierarchy. Philosophy of Science, 68, 53–74. Craver, C. F., & Darden, L. (2001). Discovering mechanisms in neurobiology: The case of spatial memory. In P. Machamer, R. Grush, & P. McLaughlin (Eds.), Theory and method in the neurosciences (pp. 112–137). Pittsburgh: University of Pittsburgh Press. Dennett, D. C. (1991). Consciousness explained. Boston: Little, Brown. Domhoff, W. G. (2003). The scientific study of dreams. Neural networks, cognitive development, and content analysis. Baltimore, MD: American Psychological Association/Port City Press. Dretske, F. (1995). Naturalizing the mind. Cambridge, MA: MIT Press. Flanagan, O. (1995). Deconstructing dreams: The spandrels of sleep. The Journal of Philosophy, 92, 5–27. Flanagan, O. (2000). Dreaming is not an adaptation. Behavioral and Brain Sciences, 23(6), 936–939. Foulkes, D. (1982). Children’s dreams: Longitudinal studies. New York: Wiley. Foulkes, D. (1985). Dreaming: A cognitive-psychological analysis. Hillsdale, NJ: Lawrence Erlbaum. Foulkes, D. (1999). Children’s dreaming and the development of consciousness. Cambridge, MA: Harvard. Franklin, M. S., & Zyphur, M. J. (2005). The role of dreams in the evolution of the human mind. Evolutionary Psychology, 3, 59–78. Hartmann, E. (1995). Making connections in a safe place. Is dreaming psychotherapy? Dreaming, 5, 213–228. Hartmann, E. (1996). Outline for a theory on the nature and functions of dreaming. Dreaming, 6, 147–170. Hartmann, E. (1998). Dreams and nightmares: The new theory on the origin and meaning of dreams. New York: Plenum Press. Hobson, J. A. (1988). The dreaming brain. New York: Basic Books. Hobson, J. A., & McCarley, R. W. (1977). The brain as a dream state generator: An activation-synthesis hypothesis of the dream process. American Journal of Psychiatry, 13, 1335–1348. Hobson, J. A., Pace-Schott, E. F., & Stickgold, R. (2000). Dreaming and the brain: Toward a cognitive neuroscience of conscious states. Behavioral and Brain Sciences, 23(6), 793–842. Humphrey, N. (2000). Dreaming as play. Behavioral and Brain Sciences, 23(6), 953. Kahn, D., & Hobson, J. A. (2005). Theory of mind in dreaming: Awareness of feelings and thoughts of others in dreams. Dreaming, 15, 48–57. Kramer, M. (1991). The nightmare: A failure in dream function. Dreaming, 1, 227–285. Kramer, M. (1993). The selective mood regulatory function of dreaming: An update and revision. In A. Moffit, M. Kramer, & R. Hoffman (Eds.), The functions of dreaming (pp. 139–195). New York: State University Press. McNamara, P. (2004). An evolutionary psychology of sleep and dreams. Westport, CT: Praeger. McNamara, P. (2007). Costly signaling theory of dream recall and dream sharing. In P. McNamara & D. Barrett (Eds.), The new science of dreaming (pp. 117–132). Westport, CT: Praeger. Nielsen, T. A. (2000). Covert REM sleep effects on REM mentation: Further methodological considerations and supporting evidence. Behavioral and Brain Sciences, 23(6), 1040–1057. Nielsen, T. A., & Germain, A. (2000). Post-traumatic nightmares as a dysfunctional state. Behavioral and Brain Sciences, 23(6), 978–979. O’Regan, J. K., & Noë, A. (2001). A sensorimotor account of vision and visual consciousness. Behavioral and Brain Sciences, 24(5), 939–1031. Pagel, J. F. (2003). Non-dreamers. Sleep Medicine, 4, 235–241. Peterson, J. B., & DeYoung, C. G. (2000). Metaphoric threat is more real than real threat. Behavioral and Brain Sciences, 23(6), 992–993. Revonsuo, A. (2000). The reinterpretation of dreams: An evolutionary hypothesis of the function of dreaming. Behavioral and Brain Sciences, 23(6), 877–901. Revonsuo, A. (2001). On the nature of explanation in neurosciences. In P. Machamer, R. Grush, & P. McLaughlin (Eds.), Theory and method in the neurosciences (pp. 45–69). Pittsburgh: University of Pittsburgh Press. Revonsuo, A. (2006). Inner presence. Consciousness as a biological phenomenon. Cambridge, MA: MIT Press. Schredl, M. (2000). Continuity between waking life and dreaming: Are all waking activities reflected equally often in dreams. Perceptual and Motor Skills, 90, 844–846. Schredl, M., & Hofmann, F. (2003). Continuity between waking activities and dream activities. Consciousness and Cognition, 12, 298–308. Searle, J. R. (1992). The rediscovery of the mind. Cambridge, MA: MIT Press. Strauch, I., & Meier, B. (1996). In search of dreams. Results of experimental dream research. New York: SUNY Press.
1090
K. Valli / Consciousness and Cognition 20 (2011) 1084–1090
Thompson, E. (2001). Empathy and consciousness. Journal of Consciousness Studies, 8(5–7), 1–32. Tinbergen, N. (1963). On aims and methods in ethology. Zeitschrift für Tierpsychologie, 20, 410–433. Tye, M. (1995). Ten problems of consciousness. A representational theory of the phenomenal mind. Cambridge, MA: MIT Press. Valli, K. (2008). Threat simulation – The function of dreaming? Doctoral dissertation. Finland: University of Turku, Turku. Wallach, M. S. (1963). Dream report and some psychological concomitants. Journal of Consulting Psychology, 27(6), 549. Wamsley, E. J., Tucker, M., Payne, J. D., Benavised, J. A., & Stickgold, R. (2010). Dreaming of a learning task is associated with enhanced sleep-dependent memory consolidation. Current Biology, 20(9), 1–6.
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Dreaming in the multilevel framework q Katja Valli ⇑ Centre for Cognitive Neuroscience, University of Turku, Finland School of Humanities and Informatics, University of Skövde, Sweden
a r t i c l e
i n f o
Article history: Available online 13 May 2011 Keywords: Biological realism Multilevel framework Dreaming Phenomenal consciousness
a b s t r a c t Biological realism (Revonsuo, 2001, 2006) states that dreaming is a biological phenomenon and therefore explainable in naturalistic terms, similar to the explanation of other biological phenomena. In the biological sciences, the structure of explanations can be described with the help of a framework called ‘multilevel explanation’. The multilevel model provides a context that assists to clarify what needs to be explained and how, and how to place different theories into the same model. Here, I will argue that the multilevel framework would be useful when we try to construct scientific explanations of dreaming. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction Dreaming is the most often naturally occurring altered states of consciousness, experienced on a nightly basis by most individuals (Pagel, 2003; Strauch & Meier, 1996; Wallach, 1963). Dreaming is thus a universal phenomenon, and the brain’s capability to produce subjective hallucinatory experiences during sleep seems to be a part of our biological machinery, a trait genetically passed on in successive generations. Nevertheless, despite rigorous empirical research and theorizing, it is still far from clear why our brain produces these experiences, and how they are created. 2. Biological realism and the multilevel framework Biological realism (Revonsuo, 2001, 2006) treats consciousness as a real natural biological phenomenon that ‘‘literally resides within the confines of the brain’’ (Revonsuo, 2006, p. 10).1 Thus, it follows that dreaming as a form of subjective consciousness can be given a fully naturalistic explanation, similar to the explanation of other biological phenomena. Consequently, the explanatory framework used in biological sciences could also be used in dream research as well as more generally in consciousness research.2 In biological sciences the underlying explanatory strategy is ‘multilevel explanation’ (e.g., Bechtel, 1994; Bechtel & Richardson, 1992, 1993; Craver, 2001; Craver & Darden, 2001; Tinbergen, 1963). Basically, biological systems are hierarchical, composed of multiple but interconnected levels of organization. The multilevel model is utilized to conceptualize a framework for explaining the phenomenon under investigation, and it functions as a context that helps to clarify what needs to be explained and how, and how to place different theories into the same model (see Fig. 1). q This article is part of a special issue of this journal on European Science Foundation EMRC Exploratory Workshop: The Dreaming Mind-Brain, Consciousness and Psychosis (Challand Saint Anselme, Italy, 25th – 28th May 2009). ⇑ Address: Centre for Cognitive Neuroscience, Department of Behavioural Sciences and Philosophy/Psychology, Assistentinkatu 7, FI-20014 University of Turku, Finland. Fax: +358 23338770. E-mail address: katval@utu.fi 1 An approach very close to biological realism is biological naturalism, proposed by Searle (1992). In philosophy, however, biological views to consciousness are in the minority, and there are a number of other contemporary approaches, with varying philosophical commitments (e.g., Chalmers, 1996; Dennett, 1991; Dretske, 1995; O’Regan & Noë, 2001; Thompson, 2001; Tye, 1995). 2 For more details on how to apply the multilevel model to consciousness research in general, see Revonsuo (2006).
1053-8100/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.concog.2011.04.004
K. Valli / Consciousness and Cognition 20 (2011) 1084–1090
1085
Fig. 1. Phenomenon in the multilevel framework. Modified from Revonsuo (2006) and Valli (2008).
In the multilevel framework, the phenomenon under investigation is first identified and described at one level of organization (Revonsuo, 2006). The aim is to give a detailed account of what the phenomenon of interest is like. The next step involves figuring out what the components constituting and contributing to the phenomenon are. The explanation of the phenomenon can be broken down into several smaller explanatory tasks; constitutive, etiological, and contextual explanations. Each of these addresses different level of explanation in the hierarchical system (Revonsuo, 2006). Constitutive (the downward-looking) explanations involve looking at smaller-scale phenomena or a lower level of organization. They aim at decomposing the phenomenon under investigation to its constituent parts, and describing the lower level mechanisms that causally contribute to the emergence of the phenomenon. Etiological (the backward-looking) explanations focus on the temporal and causal history of the phenomenon, aiming to trace backward in time the causal chains that led to the occurrence of the phenomenon. Contextual (the upward-looking) explanations, in contrast, focus on describing the role of the phenomenon at higher level of organization where it is just one element or subsystem in a complex mechanism. Each level of explanation in the hierarchical structure, forming now a subproblem in the overall explanation of the phenomenon, can then be separately investigated by a dedicated science or with a particular method (Revonsuo, 2006). When sufficient answers to research problems on all or most levels of the framework have been acquired, we can begin to build an idealized multilevel model of the phenomenon under investigation. The main function of this model is to reveal the structure and function of the phenomenon across the hierarchical levels. As in multilevel explanation the phenomenon is approached from various levels, crossing boundaries between academic disciplines, it can be the basis for the development of an interdisciplinary discussion and consensus. 3. Dreaming in the multilevel framework If we treat dream consciousness as residing on one level of organization in a biological system, then to fully understand the phenomenon, we should insert dreaming into a multilevel framework as described in Fig. 2. First, the phenomenon itself needs to be identified, defined, and described in detail. This task is the responsibility of datadriven, descriptive dream research. The methodology required to accomplish this task is the systematic sampling of dreaming from various populations, and detailed content analyses of dream data that reveal the statistical patterns of the form and content of dream experiences. We already know a great deal, but controversies remain. Even the definition of dreaming has proven to be problematic (e.g., Nielsen, 2000), and therefore the descriptions of what dreaming is like also vary. Thus, the first task of dream science is to come to a consensus on what dreaming is, and in the case this can be accomplished, formulate a description of the phenomenon that majority agree on. 3.1. The constitutive level The constitutive explanation involves decomposing dream experience into its constituent parts, and describing the lower level mechanisms that bring about dreaming or whose activity together produces the phenomenon. Thus, the downwardlooking explanations need to describe the underlying mental, cognitive, and neural nonconscious mechanisms, both
1086
K. Valli / Consciousness and Cognition 20 (2011) 1084–1090
Fig. 2. Dreaming in the multilevel framework. Modified from Valli (2008).
structural and functional, that are the substrates for the form and content of dreams. With constitutive explanations, we run first into the ‘easy problem’ of consciousness (Chalmers, 1996). The easy problem concerns the neural or cognitive correlates of consciousness, and is mainly empirical. The constitutive level explanations for dreaming involve the ‘dream generator system’ (see also Fig. 1). Currently, mainly the Random Activation Theories (RATs) (e.g. Flanagan, 1995, 2000; Hobson, 1988; Hobson & McCarley, 1977; Hobson, PaceSchott, & Stickgold, 2000) address the constitutive mechanisms of dream consciousness in detail. In brief, RATs state that the mechanisms that actively generate REM sleep state also participate in generating dreams. The REM sleep regulating mechanisms are located in the brainstem, and the sleep related activation (for example, the ponto-geniculo-occipital (PGO) – waves) leaks over to the emotional and memory networks resulting in the random activation of these systems. The activated memories and emotions are synthesized into dream images by the forebrain that attempts to make sense of them. The dream generator system, however, is probably distinct from the REM sleep generator system, as dreaming and sleep stages can occur independently of each other.3 Therefore, dreaming should not be approached by focusing merely on the physiological aspects of different sleep stages, but rather we should look for the same underlying mechanism that produces dreaming despite of differences in brain activation patterns during different phases of sleep. Thus far, the constitutive mechanisms for dreaming, as for waking consciousness, have not been found. We know neither how the brain creates dream experiences nor where this process takes place. The methods available in cognitive neuroscience, especially as technologies evolve, can contribute to the identification and localization of the dream generator mechanism in the brain, and add to the understanding of constitutive explanations of dreaming. Nevertheless, the constitutive explanations for dream consciousness should not be misidentified with the neural correlates of dream consciousness. The constitutive mechanisms must go beyond correlative relationships, and specify the underlying hierarchical components that bring about dream consciousness. With full description of the constitutive mechanisms, we might also be able to address the ‘hard problem’ of consciousness (Chalmers, 1996), that is, by explaining the performance of all the relevant neural and cognitive systems and functions underlying phenomenal consciousness, we would know why physical processes in the brain give rise to the inner presence of mind. 3.2. The etiological level An etiological explanation aims to clarify the temporal and causal history of the phenomenon, and thus, the etiological explanations can be divided into (1) the immediate or proximate causes of the phenomenon, (2) the ontogenetic or developmental causes of the phenomenon, and (3) the ultimate or evolutionary causes of the phenomenon. In sum, the
3 Or vice versa, the currently used sleep stage classification systems are too crude to pick up similarities between sleep stages that may underlie dreaming, see, e.g., the hypothesis of covert REM processes operating in NREM sleep by Nielsen (2000).
K. Valli / Consciousness and Cognition 20 (2011) 1084–1090
1087
etiological explanations track back in time the causal chain that leads to the phenomenon. In the context of dreaming, these explanations involve neural and psychological level causes, as well as evolutionary biological causes addressing the phylogeny and possible adaptive function of dreaming. 3.2.1. The proximate causes The proximate causes of dreaming in the neural level should describe the temporal activation of the dream generator system in the brain and how it produces the internal stimulation that leads to subjective dream experience. Psychological level proximate explanations should describe how the immediate cognitive level preceding causes, such as the activation of memory sources, emotional state and activity, hyperassociativity, memory consolidation, attention, stimulus incorporation, and self-awareness, participate in producing dream phenomena. The ontogenetic or developmental explanations in the context of dreaming should aim to clarify how, during individual development from an embryo to adulthood, the brain becomes wired up so that the dream generator system matures and how the mind develops cognitively and psychologically so that it gradually produces adult-like dream experiences. Finally, the ultimate explanations should reveal the phylogenetic history of dreaming, and address the question whether dreaming was selected for its adaptive function or whether it is a spandrel or a mere random effect. The majority of dream theories include descriptions of hypothesized proximate etiological level mechanisms, and some also address the developmental level. Nevertheless, the exact causal chain of dream generation is unclear. For example, RATs (Hobson, 1988; Hobson & McCarley, 1977; Hobson et al., 2000) propose that the activation of the sleep-related mechanisms triggers the random activation of the emotional and memory networks. As the most recent memory traces encoded into the long term memory are high in saliency, and thus primed in the network, the random activation of the networks leads to the activation of most recent memories. Therefore, day residues and recent events are more often represented in dreams than events long past. The randomly activated memories are then bound into dream narratives, as the forebrain is trying to interpret the internal brain activation as it normally interprets external stimulation while awake. The narrative and meaningful nature of dream experience follows, on one hand, from the attempts by the forebrain to produce a coherent experience from the chaotic inputs received from the brain stem which has led to random memory activation. On the other hand, the binding process fails often, leading to hyperassociativity and bizarre dream features. Foulkes’s (1985, 1999) Cognitive Theory acknowledges that internal brain activation during sleep is the primary source of memory activation. Thus, Foulkes’s model seems to be compatible with the RATs in the sense that dream content is constructed from randomly or semi-randomly activated memories. Similarly, as the Continuity hypothesis (CH) (Domhoff, 2003; Schredl, 2000; Schredl & Hofmann, 2003) states that dreaming reflects wake experiences, the waking experiences stored into long term memory must be activated during sleep. Whether this activation is random or organized, the CH does not specify (although the emotional concerns of the dreamer may to some degree affect the activation of memory traces), but nevertheless, it leads to an organized and coherent dream experience. CH does neither describe how or why exactly dream content is generated from the activation. The Psychological Theories (e.g., Cartwright, 1996; Hartmann, 1995, 1996, 1998; Kramer, 1991, 1993) assume that dream content is not randomly constructed, but is based on specific activation patterns of most salient memory traces. The Threat Simulation Theory (TST) (Revonsuo, 2000) also proposes that selective and organized neural and cognitive processes are required in order for dreaming to take place. Although the theories do not explicate how exactly dreams are constructed, specifically the emotional concerns of the dreamer, or the most negatively charged or most recently activated memory traces guide dream content construction. 3.2.2. The ontogenetic causes The ontogenetic explanations for dreaming should clarify how, during individual development, the brain and mind mechanisms develop and mature so that an individual begins to dream and eventually have adult-like dreams. The Cognitive Theory (Foulkes, 1982, 1985, 1999) and the TST (Revonsuo, 2000) address developmental level etiological explanations. Children’s dreaming, according to Foulkes’s (1982, 1985) laboratory studies with awakenings from NREM and REM sleep stages is different from adults’ dreaming, because it is dependent on memory and thought processes which develop with age. Thus, dreaming follows developmental trends. The developmental level explanations for dreaming put forth by the TST are that dreaming has a genetic basis, and that dreaming reliably matures over the developmental life history of the organism according to the specifications contained in its genetic code. The maturation rate and the activity level of the threat simulation system is highly dependent on the activity level of the dream production mechanisms and the level of ecologically valid threat cue input, that is, personal threat experiences. 3.2.3. The ultimate causes The ultimate or evolutionary explanations aim to elucidate how, at some point in the evolutionary history, the human brain evolved the ability to produce dreaming. When it comes to dreaming, evolutionary explanations try to figure out how the first dream generator appeared in some distant ancestors of ours, perhaps in the first humans, perhaps in the first mammals. The phylogeny of dreaming is currently unclear, and as subjective experiences based on introspective reports can only be investigated in humans, comparative studies in other mammalian species are impossible. Dreaming may, nevertheless, be present in other species. For example, the TST acknowledges that animals may have subjective experiences during sleep to the same extent that they have subjective experiences during wakefulness (Revonsuo, 2000).
1088
K. Valli / Consciousness and Cognition 20 (2011) 1084–1090
When evolutionary psychological reasoning is extended to dreaming, the following questions about the adaptive function of dreaming emerge: Is it conceivable that the process of dreaming or dreaming about particular contents was functional in the biological sense in the ancestral world - that somehow our ancestors (independent of whether dreaming emerged in human, hominid or even more distant ancestors) had better chances of surviving or reproducing because they regularly dreamed during the night, and especially because they regularly dreamed about particular things? Could dreaming have given any kind of selective advantage for our ancestors and if yes, what kind of advantage? Does dreaming fulfill the main criterion for an adaptation, in other words, does dreaming show a clear design for a specific function, and does it seem to be specifically engineered to solving a specific adaptive problem, or did it appear as a nonfunctional epiphenomenon of other sleep-related mechanisms, as a so-called spandrel? In the past ten years, several hypothetical answers to the above questions have been presented. Dreaming has been suggested to serve adaptive functions similar to mammalian play behaviors (Bulkeley, 2004; Cheyne, 2000; Humphrey, 2000; Peterson & DeYoung, 2000). In Brereton’s (2000) ‘Social mapping hypothesis’ dreaming is suggested to have rehearsed the perceptual and emotional features required by successful social mapping, for example, by simulating the self, location, and awareness of others including awareness of their internal mental states. This is proposed to have eventually led to the emergence of self-awareness. Relatedly, Kahn and Hobson (2005) emphasize that awareness of what others are thinking and feeling is a robust aspect of human consciousness, and this aspect is maintained during dreaming. Thus, even though they do not explicitly express it, they imply that awareness of the minds of others (‘‘Theory of Mind’’) during dreaming might have contributed to the ability to anticipate the intentions of others while awake. Humphrey (2000), and Nielsen and Germain (2000) have proposed that as many of the selection pressures faced by ancestral humans were posed by complex human social life, modeling human relationships and interpersonal bonds might have had adaptive value. Franklin and Zyphur’s (2005) argument is organized along the similar lines; they propose that via dreaming it is possible to practice dealing with complex social situations, because those most adept in their social environment were likely to have the best access to resources in their social group, such as mates. McNamara (2004, 2007) suggests that costly REM features affect dream content, and through producing hard to fake emotional signals, increase fitness by facilitating cooperative social interactions or courtship displays. The psychological adaptation and creativity functions assigned for dreaming (Hartmann, 1996, 1998) suggest, though in a subtle manner, that the emotional problem solving and creativity function of dreams might have fulfilled biological functions in the ancestral environment. The TST (Revonsuo, 2000), in contrast, states that dreaming about threatening situations led to enhanced threat recognition and avoidance, thus enhancing survival and offspring production. If dream consciousness has an adaptive function, then it follows that dreaming has, evolutionarily speaking, causal powers. If phenomenal level experiences promoted the (inclusive) fitness of the organism, they must have affected behavior and organism-environment interaction in such a way that fitness promoting behaviors were more likely or more effective. Behavior and organism-environment interactions, however, belong to the contextual level explanations. 3.3. The contextual level The contextual level explanations describe the role of the phenomenon at a higher level of organization. Does dreaming relate to or affect other psychological or conscious/nonconscious phenomena and how? What consequences, if any, does the dream generator system have for learning, memory, emotion regulation, perception, and ultimately, overt behavior? In the contextual level, mental phenomenona figure in the etiological explanations for observable behavior. Thus, contextual explanations tie dreaming to other subjective and psychological phenomena on a more general level, which are intra- or interindividual level, overt behavior, and organism-environment interaction. Further, in evolutionary explanations for dreaming, the contextual level explanations need to be tied to the ultimate etiological level explanations. Contextual level explanations are addressed in many dream theories. The theories assigning a biological function for dreaming all provide a contextual level explanation, as dreaming must have modified waking consciousness or behavior in some way. The specific proposed effects range from emergence of self-awareness (Brereton, 2000) to enhancement of social skills (Franklin & Zyphur, 2005; Humphrey, 2000; Nielsen & Germain, 2000) and to threat avoidance behaviors (Revonsuo, 2000). Recently, dream imagery has also been linked to more effective sleep dependent memory consolidation (Wamsley, Tucker, Payne, Benavised, & Stickgold, 2010). In addition, Foulkes (1982, 1985, 1999) suggested that visuo-spatial skills are utilized in dream construction, and correlate with the complexity of dream content. The Psychological Theories assume that the very reason for why we dream lies in the usefulness of dream imagery for the psychological well-being of the individual (Cartwright, 1996; Hartmann, 1995, 1996, 1998), or that dreaming has a mood regulatory function (Kramer, 1991, 1993). In these theories, general psychological well-being is at the contextual level. 4. Conclusions and future directions The stand point taken in this paper is that dreaming is a biological phenomenon residing in the brain, and it can thus be explained in a similar fashion as any other biological phenomenon. Therefore, the multilevel framework can be used as a metatheory, or a model, to compare and evaluate the scope, limits, and contradictions of the more specific primary theories of dreaming, and identify conflicts between, lack of explanation, and evidence for specific theories within the same
K. Valli / Consciousness and Cognition 20 (2011) 1084–1090
1089
explanatory dimension of the model. This kind of an analysis reveal that the majority of dream theories do not address all the levels of explanation explicated in the multilevel framework, and in addition, the offered explanations are sometimes contradictory. If we are ever to build an overarching explanation and agree on what dreaming is, we need to reach reconciliation between contradictory views. The multilevel framework seems to be a promising basis for dream research, inviting contributions from various academic fields, and enabling interdisciplinary approaches to the development of an overarching theory of dreaming. The main reason for using a grounded theoretical model in dream studies is that without a model, such as the multilevel framework, that guides researchers to come up with holistic explanations, we will likely end up like the blind men studying an elephant from different perspectives, coming to completely different conclusions as to what the elephant is.
Acknowledgments I wish to thank Prof. Antti Revonsuo for his valuable comments during the preparation of the manuscript. This work was supported by the Academy of Finland (project 2600016311).
References Bechtel, W. (1994). Levels of description and explanation in cognitive science. Minds and Machines, 4, 1–25. Bechtel, W., & Richardson, R. C. (1992). Emergent phenomena and complex systems. In A. Beckermann, H. Flohr, & J. Kim (Eds.), Emergence or reduction? Essays on the prospects of nonreductive physicalism (pp. 257–288). Berlin: de Gruyter. Bechtel, W., & Richardson, R. C. (1993). Discovering complexity. Decomposition and localization as strategies in scientific research. Princeton, NJ: Princeton University Press. Brereton, D. (2000). Dreaming, adaptation, and consciousness: The social mapping hypothesis. Ethos, 28, 379–409. Bulkeley, K. (2004). Dreaming is play II: Revonsuo’s threat simulation theory in ludic context. Sleep and Hypnosis, 6, 119–129. Cartwright, R. D. (1996). Dreams and adaptation to divorce. In D. Barrett (Ed.), Trauma and dreams (pp. 173–185). Cambridge, MA: Harvard University Press. Chalmers, D. J. (1996). The conscious mind. Oxford: Oxford University Press. Cheyne, J. A. (2000). Play, dreams, and simulation. Behavioral and Brain Sciences, 23(6), 918–919. Craver, C. F. (2001). Role functions, mechanisms, and hierarchy. Philosophy of Science, 68, 53–74. Craver, C. F., & Darden, L. (2001). Discovering mechanisms in neurobiology: The case of spatial memory. In P. Machamer, R. Grush, & P. McLaughlin (Eds.), Theory and method in the neurosciences (pp. 112–137). Pittsburgh: University of Pittsburgh Press. Dennett, D. C. (1991). Consciousness explained. Boston: Little, Brown. Domhoff, W. G. (2003). The scientific study of dreams. Neural networks, cognitive development, and content analysis. Baltimore, MD: American Psychological Association/Port City Press. Dretske, F. (1995). Naturalizing the mind. Cambridge, MA: MIT Press. Flanagan, O. (1995). Deconstructing dreams: The spandrels of sleep. The Journal of Philosophy, 92, 5–27. Flanagan, O. (2000). Dreaming is not an adaptation. Behavioral and Brain Sciences, 23(6), 936–939. Foulkes, D. (1982). Children’s dreams: Longitudinal studies. New York: Wiley. Foulkes, D. (1985). Dreaming: A cognitive-psychological analysis. Hillsdale, NJ: Lawrence Erlbaum. Foulkes, D. (1999). Children’s dreaming and the development of consciousness. Cambridge, MA: Harvard. Franklin, M. S., & Zyphur, M. J. (2005). The role of dreams in the evolution of the human mind. Evolutionary Psychology, 3, 59–78. Hartmann, E. (1995). Making connections in a safe place. Is dreaming psychotherapy? Dreaming, 5, 213–228. Hartmann, E. (1996). Outline for a theory on the nature and functions of dreaming. Dreaming, 6, 147–170. Hartmann, E. (1998). Dreams and nightmares: The new theory on the origin and meaning of dreams. New York: Plenum Press. Hobson, J. A. (1988). The dreaming brain. New York: Basic Books. Hobson, J. A., & McCarley, R. W. (1977). The brain as a dream state generator: An activation-synthesis hypothesis of the dream process. American Journal of Psychiatry, 13, 1335–1348. Hobson, J. A., Pace-Schott, E. F., & Stickgold, R. (2000). Dreaming and the brain: Toward a cognitive neuroscience of conscious states. Behavioral and Brain Sciences, 23(6), 793–842. Humphrey, N. (2000). Dreaming as play. Behavioral and Brain Sciences, 23(6), 953. Kahn, D., & Hobson, J. A. (2005). Theory of mind in dreaming: Awareness of feelings and thoughts of others in dreams. Dreaming, 15, 48–57. Kramer, M. (1991). The nightmare: A failure in dream function. Dreaming, 1, 227–285. Kramer, M. (1993). The selective mood regulatory function of dreaming: An update and revision. In A. Moffit, M. Kramer, & R. Hoffman (Eds.), The functions of dreaming (pp. 139–195). New York: State University Press. McNamara, P. (2004). An evolutionary psychology of sleep and dreams. Westport, CT: Praeger. McNamara, P. (2007). Costly signaling theory of dream recall and dream sharing. In P. McNamara & D. Barrett (Eds.), The new science of dreaming (pp. 117–132). Westport, CT: Praeger. Nielsen, T. A. (2000). Covert REM sleep effects on REM mentation: Further methodological considerations and supporting evidence. Behavioral and Brain Sciences, 23(6), 1040–1057. Nielsen, T. A., & Germain, A. (2000). Post-traumatic nightmares as a dysfunctional state. Behavioral and Brain Sciences, 23(6), 978–979. O’Regan, J. K., & Noë, A. (2001). A sensorimotor account of vision and visual consciousness. Behavioral and Brain Sciences, 24(5), 939–1031. Pagel, J. F. (2003). Non-dreamers. Sleep Medicine, 4, 235–241. Peterson, J. B., & DeYoung, C. G. (2000). Metaphoric threat is more real than real threat. Behavioral and Brain Sciences, 23(6), 992–993. Revonsuo, A. (2000). The reinterpretation of dreams: An evolutionary hypothesis of the function of dreaming. Behavioral and Brain Sciences, 23(6), 877–901. Revonsuo, A. (2001). On the nature of explanation in neurosciences. In P. Machamer, R. Grush, & P. McLaughlin (Eds.), Theory and method in the neurosciences (pp. 45–69). Pittsburgh: University of Pittsburgh Press. Revonsuo, A. (2006). Inner presence. Consciousness as a biological phenomenon. Cambridge, MA: MIT Press. Schredl, M. (2000). Continuity between waking life and dreaming: Are all waking activities reflected equally often in dreams. Perceptual and Motor Skills, 90, 844–846. Schredl, M., & Hofmann, F. (2003). Continuity between waking activities and dream activities. Consciousness and Cognition, 12, 298–308. Searle, J. R. (1992). The rediscovery of the mind. Cambridge, MA: MIT Press. Strauch, I., & Meier, B. (1996). In search of dreams. Results of experimental dream research. New York: SUNY Press.
1090
K. Valli / Consciousness and Cognition 20 (2011) 1084–1090
Thompson, E. (2001). Empathy and consciousness. Journal of Consciousness Studies, 8(5–7), 1–32. Tinbergen, N. (1963). On aims and methods in ethology. Zeitschrift für Tierpsychologie, 20, 410–433. Tye, M. (1995). Ten problems of consciousness. A representational theory of the phenomenal mind. Cambridge, MA: MIT Press. Valli, K. (2008). Threat simulation – The function of dreaming? Doctoral dissertation. Finland: University of Turku, Turku. Wallach, M. S. (1963). Dream report and some psychological concomitants. Journal of Consulting Psychology, 27(6), 549. Wamsley, E. J., Tucker, M., Payne, J. D., Benavised, J. A., & Stickgold, R. (2010). Dreaming of a learning task is associated with enhanced sleep-dependent memory consolidation. Current Biology, 20(9), 1–6.