Preprint of an article published in the journal Complexity, 6(1):20-26 (2000) [Wiley URL]

The Complex Link Between Neuroanatomy and Consciousness

Giorgio A. Ascoli

Krasnow Institute for Advanced Study and Department of Psychology
George Mason University, MS2A1, 4400 University Dr. - Fairfax, VA 22030-4444
Ph. (703) 993-4383, Fax (703) 993-4325, e-mail: ascoli@gmu.edu



Abstract: This essay is a commentary on Edelman and Tononi's A Universe of Consciousness – How Matter Becomes Imagination, a scientific book on the mind-body relationship. The underlying theme of the book is that consciousness is a process emerging from highly integrated and distributed activity in the thalamocortical system in the brain (the “dynamic core”). The authors use this hypothesis to explain several fundamental characteristics of consciousness in terms of an interesting definition of neural complexity. A Universe of Consciousness presents a theory well-grounded in neuroanatomy, and succeeds in connecting neurobiological knowledge with subjective experience. The result is a convincing story that unfolds from an accurate description of the physical and biological processes in the nervous system (the first two thirds of the book) to an intriguing explanation of qualia, the private, qualitative, and subjective aspects of consciousness. After reviewing the book, we comment on specific scientific and philosophical aspects of Edelman and Tononi's approach. Particularly, we discuss whether the implications of the “dynamic core” hypothesis involve the establishment of new physical laws or can be reduced to presently accepted scientific theories. In addition, we explore the consequences of establishing the primacy of neuroanatomy in the study of consciousness with a thought experiment on qualia.

In A Universe of Consciousness (Basic Books, NY, 2000), Gerald Edelman and Giulio Tononi face the mind-body problem straight up, guiding the readers hand-by-hand through a fascinating and exciting journey of 222 intense pages. The subject of consciousness (“How Matter Becomes Imagination” is the subtitle of the volume) is arguably the hardest and most debated issue in the history of philosophy. The authors confidently take up the challenge shielded by a solid framework of neurobiology and brandishing sharp and original ideas. They emerge as winners with an excellent book that will become a classic in the field. Edelman is not new to this arena: his “The Remembered Present: A Biological Theory of Consciousness”, published in 1989 also by Basic Books, explores several of the ideas that are used in A Universe of Consciousness. This new book, however, is explicitly and uniquely aimed at the very core of consciousness, i.e. the raw feelings of awareness, or qualia.

A Universe of Consciousness is a true pleasure to read, thanks to a clear and enjoyable writing style and an excellent organization of the ideas. The arguments are neatly laid out, with a good balance of everyday metaphors and accurate scientific descriptions, and a catchy use of mild rhetoric. The most technical and subtle aspects of the discussion are left to notes at the end of the book, which are complemented by the bibliographic references. As a consequence, the book is accessible at many levels, from the non-expert to the academic. The book is divided in six parts and seventeen chapters. Each part is introduced by a one-page explanation of the basic terminology and the specific aims of the proposed ideas. Additionally, every single chapter is preceded by a summary or “abstract” (in the style of scientific articles). Those readers for whom this book is the first introduction to the themes of consciousness and/or neurobiology may benefit from a preliminary reading of the preface, the introductions to the Parts, and the summaries to the Chapters before taking up the complete book.

Part I constitutes a simple, yet enlightening, introduction to the problem of consciousness and the philosophical paradox of subjective experience. How can the sensations of color, temperature, pain, touch, love, etc. arise from physical processes? No matter how detailed and complete our neurobiological knowledge is, will it “explain” why private phenomenological qualities (qualia) are the way we experience them? Should we even bother studying the nervous system if we are interested in consciousness? Many attempts have been made in recent philosophical literature to define and solve this problem (for a collection, see e.g. [1]), i.e. the “explanatory gap” involving qualia [2]. In analyzing this issue, philosophers have reached varied and diverse conclusions, from mysterianism [3] (but see also [4]) to reductionism [5] and calls for a quantum theory of consciousness (e.g. [6]). Edelman and Tononi avoid a long-winded reworking of ideas well-represented in the literature; they skim through classic dualism and materialism to show the reader what the overall problem is, and then move on to build a solution.

The second Part constitutes an excellent survey of basic neurobiology. From system neuroanatomy to the neurotransmitter level, Edelman and Tononi do an outstanding job explaining all the key concepts to be used through the rest of the book, keeping jargon to a minimum and defining non-trivial terms. At the same time, they carefully avoid diverging into neurobiological technicalities and information that, although fascinating and scientifically crucial, are nonetheless not essential to the construction of their theory of consciousness. Such a choice makes this book more compact, clear, enjoyable, and convincing than many other recent volumes on consciousness and science that attempt to give an omni-comprehensive view of physics and/or biology. This Part of the book outlines the primacy of neuroanatomy and the concept of “reentry”, a thick system of parallel, mutual feedback between brain regions or among subparts of a brain region. The authors also introduce crucial notions that seem to create a logical correspondence between brain activity and consciousness, such as coherence, integration, and differentiation. These important concepts are repeated and developed further in the subsequent chapters.

Part III is a discussion of the Darwinian perspective on neurobiology, which is probably Edelman's foremost contribution to the field of neuroscience. The authors outline the theory of neuronal group selection (or TNGS), and claim that natural neural networks that have evolved by mutation and competition are computationally more powerful than standard (serial or parallel) algorithm-performing computers. In nature, neuronal group selection acts in different ways, including the development of the nervous system, experience acquisition, and the fitting of the best neuronal patterns to store an association or perform an action. The chapter on nonrepresentational memory attempts to describe memory as the ability of a dynamic system to repeat or suppress a mental or physical act, without the need to store information. Unlike the rest of the book, the explanations in this section are cursory, and the reader is left without understanding the issue as deeply as other topics. The discussion on symbols, concepts, and neuronal “representation” is too narrow and superficial to be fully convincing. Nevertheless, these concepts constitute a good transition to the exposition on the “remembered present”, which provides a link between associative memory, abstraction, and perception (see also [7] and [8]). The authors also highlight the crucial contribution of “value” systems, i.e. subcortical reinforcers that drive selection and memory towards the best fit of neuronal groups and cognitive functions.

Part IV revisits the fundamental issues of reentry, synchrony, coherence, and differentiation from a quantitative perspective. The authors describe the approach they developed with their colleague Olaf Sporns to define neuroanatomical measurements, which, from a theoretical standpoint, are crucial in capturing functional processing in the nervous system. The key definition is that of “mutual information” between a subset of a system and the rest of the system, expressed as the difference between the entropy of the whole system and the sum of entropies of the subset and of its complement. In other words, mutual information measures the total amount of statistical dependence between a subset and its complement. The importance of this concept is that it does not rely on the existence of an “external” observer to define information. This approach bypasses the “homunculus” problem that would plague any attempt to use Shannon’s more standard definition of information [9, 10] in the context of perception and consciousness. Edelman and Tononi convincingly argue that mutual information measures the “difference that makes a difference”, i.e. the information available in (a subset of) the nervous system, that can be detected within the nervous system itself. The authors then expand this concept by defining “neural complexity” as the average mutual information of all possible bipartition of the system, and by showing with extremely clear examples the striking correspondence between this definition and the common use of the term “complexity” by neuroscientists.

Part IV also introduces two other definitions to quantify precisely the interactions among the elements of a system (“integration”) and the functional grouping of a subset of elements in a system (“cluster index”). Since the notions of information, complexity, integration, and functional cluster are the cornerstones of Edelman and Tononi’s theory of consciousness, these definitions represent an invaluable reference for the remainder of the book. The particular care, precision and consistency of these definitions (formulated with simple mathematical equations) make the subsequent discussion extremely solid and unambiguous.

The last portion of Part IV constitute a bridge to the following, central section of the book, and deals with the pivotal problem of the “neural correlates of consciousness” [11], or, as the authors put it, “where the knot is tied”. The previous chapters of the book had already placed the thalamocortical system at center stage, as suggested also by Baars [12], Taylor [13] and Singer [14], among others. The last chapter of Part IV summarizes the evidence that consciousness does not require the entire body, nervous system, or brain to be sustained, but just a portion of the brain. However, the authors note that the neuronal groups underlying conscious experience, called the “dynamic core”, are widely distributed and integrated. The dynamic core is not coincident with the entire thalamocortical system, but largely overlapping with it. The key hypotheses defining the dynamic core are based on the concepts developed at the beginning of this Part of the book: the dynamic core is a distributed, highly differentiated (i.e., highly complex) functional cluster achieving high integration in hundreds of milliseconds. The thalamocortical system offers the right potential for high integration and differentiation thanks to an extensive network of reentrant connections. The authors persuasively maintain that some of the key cognitive aspects of consciousness (privateness, limited capacity, integration, differentiation) can be explained in terms of corresponding properties of the dynamic core, ultimately due to its complexity. Additionally, they show that reentry in the thalamocortical system is the key neuroanatomical substrate for these processes.

Part V represents, by all means, the very apex of A Universe of Consciousness. By this point, the problem of qualia has been laid out clearly. The discussion of the neuroanatomical foundations of the physical processes underlying consciousness culminate in the key question of the entire book: exactly what process goes on in the brain when we “feel” a quale, and why do we feel it when that process is active? In order to answer this question, Edelman and Tononi use the example of color qualia. After a concise but lucid description of basic color psychophysics, they consider the very subjective aspect of color sensation. The authors refute the “atomistic” idea that the quale of red would correspond to a single neuronal group “R” (even under the assumption that R would be active if and only if redness is felt). In contrast, they submit the hypothesis that the N-dimensional space of the dynamic core (where N would be the large number of neuronal groups making up the dynamic core) should be viewed as a whole. The pure sensation of red would thus correspond to an entire state of the dynamic core, including the activity of R, but also the lack of activity of all other groups.

The dynamic core hypothesis and the concept that qualia correspond to an entire state of a high-dimensional space, naturally complement the idea that the “information that counts”, in the nervous system, consists of the ability to differentiate among a large number of states with a fast (~0.1 seconds) integration. Part V introduces several intriguing ideas which emerge from this discussion. First, as points in the phase space of the dynamic core, qualia obey a certain metric. Intuitively, the integrated quale of a red triangle will be somehow related to the quale of red. By the same token, yellow and red will be closer than, say, sweet and red. In this metric, one can barely imagine the complex mental relationships underlying, for example, the sensations corresponding to the thought itself of complexity. Needless to say, the definitions of complexity and information proposed by Edelman and Tononi constitutes an ideal starting point for quantitatively defining the metric of qualia. Second, as the thalamocortical system is constantly reshaped by experience and neural activity, so is the space of qualia (thus allowing individuals, for example, to learn the distinctive tastes of different wines). Only some aspects of neural representation, however, are even possibly accessible to conscious perception: blood pressure, for instance, although carefully controlled by the nervous system, is not part of the qualia space. The circuits underlying blood pressure detection and control, in fact, are not connected through reentry with the thalamocortical dynamic core. A third, brilliant subject of discussion in Part V explores the implication of this framework for our understanding of the relationship between the conscious and the unconscious. Through a neuroanatomical analysis of the parallel appendages to the thalamocortical system, particularly basal ganglia, but also cerebellum and hippocampus, Edelman and Tononi depict a scenario of ports in and out connecting the conscious core to the rest of the nervous system, body, and world. This description nicely accounts for how we consciously learn a routine (e.g., to play a piece of music on a piano) and how, through practice, this routine becomes unconscious (and more efficient).

Until this point in the book, A Universe of Consciousness is logically constructed, step by step, starting from sound hypotheses, factual neurobiological knowledge, original ideas, and much of the available experimental evidence from cognitive science, neuroanatomy, and neurophysiology. Part VI attempts to build further on the proposed theory and to include explanations of high-order consciousness (being conscious of being conscious). However, despite the importance of these issues, Part VI lacks the supporting arguments to deliver a convincing story. The discussion on language and the self simply fails to reach the uncontroversiality of the rest of the book, and presents the authors’ opinions without thorough scientific justification. Edelman and Tononi claim that self-consciousness, but also symbolic memory, fictious imagination, and even thought, require language. Unfortunately, they neglect to define the boundaries of their conclusions and, most importantly, to discuss much experimental evidence involving higher animal communication (e.g., [15-17]). These 20 pages would have been put more clearly in context if presented as a speculative appendix to the main body of the book. Without this distinction, the authors risk diluting some of the more interesting intellectual goals achieved in the challenge of explaining qualia and defining neural information.

Throughout Edelman and Tononi's book, one is invited to reflect on several scientific and philosophical conclusions that are supported by neurobiological evidence and by the authors' reasoning. Based on the Jamesian conviction that consciousness is a process, not a thing, the scheme emerging from the dynamic core theory relates the key characteristics of consciousness (e.g. coherence, unity, and integration) with corresponding properties of the thalamocortical system. These properties are ultimately due to the extensively reentrant neuroanatomical connections in the cortex, thalamus, and related structures. In their analysis, however, the authors go beyond the classical search for “the neural correlates of consciousness”, because they maintain focus on the process, rather than on the location in the brain. They are more interested in the functional properties of thalamocortical neuroanatomy than on the very specific substructures, and rightly so. At page 42, one reads “[...] the single word most significant for understanding the brain [is] neuroanatomy” (emphasis theirs). Using similar words, we should say that the single conclusion most significant in A Universe of Consciousness is that neuroanatomy plays an absolutely pivotal role in the process of consciousness. Thus, if we are to fully understand consciousness, it is from neuroanatomy that we should start.

Edelman and Tononi put their own research efforts where their mouths are, and go a long way to lay the neuroanatomical foundations of their theory. They describe the pilot experimental studies of brain imaging on human subjects, suggesting that similar or identical conscious experiences in different individuals are likely to be correlated to different neural processes and even different locations within the thalamocortical system. In other words, the neural correlates of consciousness are as private as consciousness itself. They break new ground in theoretical neuroanatomy defining units to measure integration, functional clustering, mutual information, and neural complexity. They report results of computer simulations indicating that these statistical definitions reflect the more intuitive properties of integration, coherence, and differentiation from the neurobiological and cognitive standpoint. So defined, these units are useful to quantitatively characterize the neuroanatomical changes through development. Finally, they prove that the physical properties of the processes emerging from a highly reentrant network such as the thalamocortical system, i.e. high complexity and high integration, closely correspond to the cognitive properties of the process of consciousness, i.e. high differentiation and high informativeness.

Undoubtedly, A Universe of Consciousness brings novel ideas to the scientific discussion on consciousness. The Theory of Neuronal Group Selections provides a fertile framework to understand memory and higher cognitive functions. Using the TNGS, Edelman and Tononi dive into the details of specific sensations and automatic processes. They often borrow neurological examples in support of their theses, discussing the physiological and pathological states of altered consciousness, such as sleep, anesthesia, epilepsy, schizophrenia, personality disorders, and “split brain”. Overall, the authors do a superb job in building a coherent story from neuroanatomy to cognition.

If a fault needs to be found in this book, it is the lack of emphasis on the bridge that is postulated to connect “objective” neuroanatomy with “subjective” cognition. Can the existence of qualia, and all their properties, be logically deduced from neuroanatomical data? In Edelman and Tononi's theory, this question is relegated somewhat ambiguously to the introduction of the book, and it is not acknowledged where the more philosophically inclined reader would expect the punchline. The authors focus extensively on the properties of the neural process (integrated activity in the highly reentrant dynamic core) and on the properties of consciousness (unity, privateness, coherence, informativeness), but they only briefly state their view on the connection between a specific neural state and a specific quale:

“The pure sensation of red is a particular neural state identified by a point within the N-dimensional neural space defined by the integrated activity of all the groups of neurons that constitute the dynamic core. The quale of the pure sensation of red corresponds to the discrimination that has been made among billions of other states within the same neural reference space. While neurons responding to the presence of red are certainly necessary for the conscious experience of red, they are clearly not sufficient. The conscious discrimination corresponding to the quale of seeing red acquire its full meaning only when considered in the appropriate, much larger neural reference space. By this same argument, if the same neuronal groups responding to red were firing precisely in the same way but were functionally disconnected from the core, such firing would have no meaning and no associated quale” (p. 167)

The above paragraph is extremely clear, and, relying on two thirds of the book that seem to prepare the ground specifically for this connection, the statement is sound and compelling. However, the authors do not emphasize that this is indeed their closure of the gap, the very bridge connecting mind and body in their theory. As a consequence, the reader is left with no new insight with respect to the distinction between postulated connection and logical connection. Is there a reason why a sensation corresponds to a specific state of the dynamic core (as opposed to another one)?  The book does not address this question, though the authors deny that the law of correspondence between qualia and points in the space of neuronal activity in the dynamic core is a “new” physical law. They claim that “[...] only conventional physical processes are required for a satisfactory explanation of consciousness - no dualism is allowed” (p.14), and subsequently, after equating qualia to meaning: “It is the amazingly complex material structures of the nervous system and body that give rise to dynamic mental processes and to meaning. Nothing else need to be assumed - neither other worlds, or spirits, or remarkable forces as yet unplumbed, such as quantum gravity” (p.219).

One of the most debated issues in the recent philosophical and scientific discussions about consciousness concerns the (ir)reducibility of qualia to known physical laws. Is the physical law ascribing a quale to a particular state of the dynamic core reducible? If so, how? David Chalmers proposed a distinction between an “easy” problem and a “hard” problem in the study of consciousness [18]. The “easy” component (which is only epistemologically easy!) consists of characterizing the chemical structures and the physical processes that appear to be (empirically) necessary, and possibly sufficient, to give rise to consciousness. The “hard” phase of the study (which would actually be quite easy once the “easy” problem is solved) will consist of “admitting” that there is no “logical” reason for those chemical structures and physical processes to be consciousness: rather, they will correspond to consciousness, or give rise to it. There will be no choice, Chalmers argues, but to elevate such a correspondence to the status of irreducible, or fundamental, physical law. Correspondingly, qualia, or at least some “fundamental” qualia will be considered new “elementary” constituents of the Universe and of Reality [19].

Whether or not one agrees with Chalmers' analysis, it is common epistemological knowledge that natural laws, and natural entities, can be reducible or irreducible. For example, thermodynamics can be reduced to statistical mechanics, optics to electromagnetism, and genetics to molecular biology. A reductive step need not be the last one (for example, molecular biology can be reduced to chemistry, which can be reduced to quantum mechanics). However, it has to fully explain a phenomenon with already accepted physical laws (which usually have been previously introduced to describe a different phenomenon). For example, when we claim that the temperature of a gas is reducible to the average velocity of its molecules, we prove that all the observable, objective properties of temperature, as well as its definition, are a direct logical consequence of Newton's laws of mechanics. Let us imagine a scientist, arriving from another Universe (ruled by different physical laws), to whom we explain all the laws of mechanics, but not those of thermodynamics. Suppose we tell this extraterrestrial scientist about a state function of gas, called temperature, which corresponds to the average velocity of the gas molecules, and then ask “Does energy pass from a high-temperature body to a low-temperature body, or vice versa? Is temperature proportional to the product of volume times pressure, or to their ratio?”. The fact that thermodynamics is reducible to mechanics means that, out of sheer logic, and without carrying out experiments, the scientist will be able to provide the correct answers using only the laws of mechanics. No additional physical law is needed to encapsulate thermodynamics.

In contrast, the fact that two bodies attract each other with a force inversely proportional to the square of their distance, cannot apparently be reduced to other physical laws. We simply cannot explain the properties of gravitation logically; we can only describe them. Suppose we explained all our physics except Newton’s law of gravitation to the above alien scientist and then mentioned the existence of an attractive force between any two bodies, called gravitation. If we then asked “is gravitation inversely proportional to the square of the distance, or to its cube?”, there would be no way for the scientist to deduce the answer logically. In order to account for gravitation, we need a new, additional law (and, as a matter of fact, an additional entity such as gravitational mass). An irreducible law need not remain irreducible forever. Electricity and magnetism were considered irreducible entities, until they were recognized as a single phenomenon and reduced to the properties of subatomic particles. At any given time, the frontiers of Science consist of a set of irreducible laws and fundamental entities. Reductive steps are always great advances, but in the end, when all the natural phenomena are “explained”, all the knowledge will still be based on one or more fundamental laws and entities.

Whoever admits the existence of subjective sensations, feelings, emotions, in other words, qualia, and seeks to frame them as observables within a physical or biological theory, must address the issue of their (ir)reducibility. Either all the properties of qualia are explained logically in terms of the chemical structures and physical processes to which they correspond, or the law of correspondence should be deemed irreducible. Edelman and Tononi endorse this distinction: “[…] we examine what kind of neural processes actually explain the fundamental properties of consciousness, rather than merely correlate with them” (p. 19, emphasis theirs). The dynamic core theory does explain why the subjective experience emerging from an integrated and differentiated neural process is unitary and informative. However, the subjective sensation of a conscious state is also a fundamental property of that state (in fact, it is arguably its most fundamental property!), and it is the explanation of this property in terms of neural processes that eludes current philosophers and scientists. The authors set their own challenge: “No matter how accurate the description of the physical process underlying it, it is hard to conceive how the world of subjective experience - the seeing of blue and the feeling of warmth - springs out of mere physical events” (p. 2). However, in A Universe of Consciousness they do not explain how their theory allows an objective observer to deduce which quale correspond to any given neural state, and vice versa; nor do they admit that it is not possible to deduce such a correspondence logically. For example, can the very subjective sensation of red be explained from the corresponding point in the neural space by other natural laws, or does this correlation need to be postulated to account for the experimental reality? The reader suspects that  the correspondence law proposed by Edelman and Tononi is, for the moment, “irreducible” or fundamental, despite the authors’ protest that they have explained qualia without invoking new physical laws.

This ambiguity in the dynamic core theory is only a minor problem, and does not affect the excellence of both content and form of Edelman and Tononi's ideas on qualia. It is, however, reflected in some (also minor) confusion, diffused through the book, regarding the fundamental properties of consciousness and what exactly the book aims to explain. The authors insist on key characteristics of conscious experience, namely integration or unity, differentiation or informativeness, privateness, coherence, limited capacity, flexibility and plasticity. These are certainly necessary conditions for an experience to be called “conscious”. But are they also sufficient? A Universe of Consciousness does not specify which properties must make a neural process conscious; the focus is kept only on the conditions that may give rise to consciousness. For example, at page 14 one can read a sort of “specific aims” of the book: “[...] to provide a satisfactory scientific account of consciousness [...]: what kind of physical process it is, why it has the properties it has, and under what conditions it may occur”. The declaration of intents later becomes even more restrictive: “We do not attempt to explain everything [...]. Instead, we concentrate on certain fundamental properties of consciousness” (p. 18), which turn out to be, precisely, integration, differentiation, coherence etc. The authors never address the question of whether there are neural processes which display the above properties but remain, nonetheless, unconscious.

Despite this lack of emphasis, A Universe of Consciousness is not limited to the “easy” problem, as delineated by Chalmers. The authors do indeed address the “hard problem” (without recognizing the “hardness”), and the correspondence between qualia and states of the dynamic core proposed in this book actually yields several “hard” conclusions. For example, the claim that the state of the entire dynamic core (rather than that of just a few neuronal groups) is crucial to define a quale, refutes the “modular” approach (“one neuronal group, one quale”) recently suggested [20, 21]. At the same time, the localization of the key properties of the process within the thalamus, cortex, and related areas, goes against the view that consciousness is distributed through the entire brain [22] or through the whole nervous system, body, and environment [23]. Most importantly, the solution to the hard problem constituted by the dynamic core theory establishes a principle of “qualia egalitarianism”: each quale (subjectively complex or simple; abstract or emotional) is a point in the neural space of the integrated activity in the reentrant networks of the thalamocortical system. With the definitions of mutual information and neural complexity, a first approximation of qualia metrics emerges from this theory, and the discrimination between conscious and unconscious information is straightforward.

Edelman and Tononi's theory of qualia is biologically convincing and, thanks to its neuroanatomical foundations, lends itself to interesting thought experiments. One of the most quoted sentences in the modern philosophical debate around consciousness is Nagel's “what is it like to be a bat?” [24]. This questions is intriguing not only because the words “what it is like to be” perfectly capture the meaning of subjectivity, but also because we as humans do not have the sense of echolocation, and so the answer seems hard even to imagine. A Universe of Consciousness does not tackle Nagel's question. However, the positions supported in this book suggest an experiment that, at least in principle, could provide an (indirect) answer. Imagine that we perform the following surgical intervention on a human being (since this is a thought experiment, we need not address any ethical or technical issues): within a small fraction of skin surface (say, a few square millimeters on the forehead), we detach all the touch nerve endings from the mechanoreceptors of the skin and we hook them up to an external machine. Now, when the machine is “activated”, at least at the beginning, the subject could “feel” touched in the corresponding position of the forehead. Possibly, when physically touched in that position, the subject would not feel anything (we say possibly because, if the affected surface is kept to a minimum, the nervous system is likely to “fill in” this “somatosensory blind spot”). Let us then imagine doing the same experiment for the other sensory modalities: we disconnect the retinal photoreceptors in a small position of the visual field, the auditory receptor selective for a small range of frequencies, etc. All the disconnected sensory nerves are connected to the same “machine”. Now let us suppose that this machine is continuously emitting ultrasounds to the environment and its sensory “activity” depends on the reception of the ultrasounds back. In other words, we have installed a low-frequency radar on our subject, and different patterns of echolocation are transduced in his/her nervous system as firing patterns of a neuronal group patched together from neurons formerly belonging to different sensory systems.

The question is: will this surgically modified human being acquire, with experience, (some of) the qualia of echolocation? If so, (s)he could try and describe to us what it is like to be a bat (or at least some aspects of it). If nothing else, one of us humans would know, and everyone (philosophers, neuroscientists, and individuals in search of new emotions) would have the option to undergo the same operation and feel for themselves. According to the neuroanatomical criteria described in A Universe of Consciousness, since all of the signaling pathways connected to the radar would have access to the dynamic core, conscious experience might in fact arise from echolocation, provided that the neuronal group responding to the radar signals in the thalamocortical system could form a functional cluster achieving, through reentrant interactions, high integration in hundreds of milliseconds. If it is doubted that an adult nervous system would be plastic enough to rearrange its connection to account for the “unitarity” of the new experience, we can still think of the same procedure carried out on a very young subject, or even on a fetus.

Perhaps to settle conclusively the issue of the necessary and sufficient conditions to obtain consciousness, we will indeed need to revert to experiments with artificial sensory organs. One very intriguing subject of discussion then regards the possibility that entirely artificial devices (robots) might one day claim conscious experience. Edelman and Tononi decide to skim over this topic, despite leading one of the most advanced research laboratories in this field worldwide. Their logical framework, based on the  neuroanatomical properties of the thalamocortical system, is fertile ground for a deep and original scientific discussion on artificial consciousness. Edelman’s own idea of “the remembered present” as well as the contributions of other authors to the recent literature [4, 7, 8, 17, 25, 26] might have allowed a “bottom-up” approach to artificial consciousness in this book. The authors unfortunately do not embark on such theoretical implications. Instead they provide an analysis of higher-order consciousness, language, symbols, concepts, and imagination.

A Universe of Consciousness is clearly not a review of the many discussions and ideas on consciousness that have flourished in the past decade (not does it pretend to be). Rather, it is a self-contained exposition of a theory that, although not radical or revolutionary, is sound, insightful, and almost entirely built on previous original ideas by the same authors. All the key concepts are elaborated extensively, while little is mentioned about alternative theories of consciousness or other related issues such as free will. Accordingly, the bibliography is neat and pertinent, but certainly not exhaustive for the field of consciousness studies.

In conclusion, A Universe of Consciousness is highly recommendable. Many volumes, articles, journals, and books on consciousness have been offered to the public in the past decade, promising to say the last word on the mind-body relationship. Some of these were purely philosophical treatises with no foundation in biology. Others accurately and massively summarized human knowledge, from physics to natural science, but hardly dealt with the key issue of qualia. Some, to be fair, did attempt to connect neurobiology with cognitive philosophy, but very few, if any, were able to tell a whole and coherent story, from the complexity of neuroanatomy, to the very core of consciousness. Finally, from this score of interesting but never completely satisfying publications, emerges a book that is compact, yet complete, profound, yet easy to read. A Universe of Consciousness is original and stimulating, as only truly novel theories can be, but it reaches peaks of clarity and maturity that are only usually found in popular reviews. This is the greatest merit for a book about the very frontier of intellect.

Acknowledgments: I gratefully acknowledge the hospitality of the Scuola Normale Superiore of Pisa, Italy, where parts of this article were written. I am also indebted to Dr. Rebecca Goldin for critically reviewing this manuscript.

References

1. T. Metzinger. Conscious Experience. Thoverton: Imprint Academic, 1995.
2. J. Levine. Materialism and Qualia: The Explanatory Gap. Pacific Phil. Quart. 64: 1983, pp. 354-361.
3. C. McGinn. The Mysterious Flame - Conscious Minds in a Material World. New York: Basic Books, 1999.
4. G.A. Ascoli. Is it already time to give up on a science of consciousness? A commentary on mysterianism. Complexity 5(1): 1999, pp. 25-34.
5. P.S. Churchland: Can neurobiology teach us anything about consciousness? In H.J. Morowitz, J.L. Singer (eds.): The Mind, the Brain, and Complex Adaptive Systems. Santa Fe Inst. Proc. Sci. Complexity 22: pp. 99-121, Reading: Addison-Wesley, 1995.
6. H.P. Stapp. Mind, Matter, and Quantum Mechanics. Berlin: Springer-Verlag, 1993.
7. D.L. Alkon: Memory’s Voice – Deciphering the Min-Brain Code. New York: HarperCollins, 1992.
8. G.A. Ascoli: Association, abstraction, and the emergence of the Self. Noetic J. 2(1): pp. 9-20, 1999.
9. C.E. Shannon, W. Weaver. The Mathematical Theory of Communication. Urbana: Univ. Chicago Press, 1964.
10. D.S. Roberston. Algorithmic Information Theory, Free Will, and the Turing Test. Complexity 4(3): 1999, pp.25-34.
11. J. Newman. Putting the puzzle together: Towards a general theory of the neural correlates of consciousness. J.Consciousness Studies 4(1&2): 1997, pp. 47-66, 101-121.
12. B.J. Baars. In the Theater of Consciousness, The Workspace of the Mind, Oxford University Press, New York, 1996.
13. J.G. Taylor. The Race for Consciousness, MIT Press, Cambridge, 1998.
14. W. Singer. Consciousness and the structure of neuronal representation. Phil. Trans. R. Soc. B 353: 1998, pp. 1829-1840.
15. G. Gallup. Can Animals Empathize? Yes. Sci. Am. Nov. 1998, http://www.scientificamerican.com/1998/1198intelligence/1198gallup.html
16. F. de Waals. Chimpanzee Politics : Power and Sex Among Apes, Johns Hopkins Univ. Pr., 2000.
17. I.M. Pepperberg, M.R. Willner, L.B. Gravitz. Development of Piagetian object permanence in a grey parrot (Psittacus erithacus). J. Comp. Psychol. 111(1): 1997, pp. 63-75.
18. D.J. Chalmers. Facing Up to the Problem of Consciousness. J. Consciounsess Studies 2: 1995, pp. 200-219.
19. D.J. Chalmers. The Conscious Mind. New York: Oxford University Press, 1996
20. F. Crick, C. Koch. Are we Aware of Neural Activity in Primary Visual Cortex? Nature 375: 1995, pp. 121-123.
21. S. Zeki, A. Bartels. The Asynchrony of Consciousness. Proc. R. Soc. B 265: 1998, pp. 1583-1585.
22. D.C. Dennet. Consciousness Explained. Boston: Little, Brown and Co., 1991.
23. M. Sheets-Johnstone. The Primacy of Movement. Philadelphia: J. Benjamin Publishing Co., 1999.
24. T. Nagel. What is it like to be a bat? Phil. Rev. 4: 1974, pp. 435-450.
25. G.M. Edelman: The Remembered Present: A Biological Theory of Consciousness. New York: Basic Books, 1989.
26. G.A. Ascoli. Progress and perspectives in computational neuroanatomy. Anatom. Rec. 257(6): 1999. pp. 195-207.