Nulla di più grande. Dalla veglia al sonno, dal coma al sogno. Il segreto della coscienza e la sua misura

by Marcello Massimini, Giulio Tononi

Adesso avete le vertigini; vertigini più forti di quelle che avete provato quando da bambini, prima di addormentarvi, pensavate, con tutta l’intensità di cui eravate capaci, all’infinità del cielo e al numero delle stelle. Forse, più forti di quelle che hanno ubriacato gli astronauti nell’istante in cui hanno visto tramontare il nostro piccolo pianeta dietro la luna. Fermarsi a soppesare il cervello in una mano è un’esperienza dirompente che dissolve, in un istante, il peso delle consuetudini, delle posizioni filosofiche, delle definizioni e delle trappole logiche che ci allontanano dal mistero della coscienza. Una domanda, urgente e semplice, sembra venire direttamente dai sensi, dai nervi della mano che soppesa: che cos’ha di diverso, di speciale, questo oggetto?

Siamo arrivati a vedere stelle lontanissime, ma non abbiamo ancora un metodo affidabile per vedere dove brilla la luce della coscienza nel mondo intorno a noi. Perché la coscienza sparisce quando ci addormentiamo per poi ricomparire nel buio della notte, più vivida che mai, durante un sogno? è cosciente un uomo che esce dal coma, ma rimane immobile e muto per mesi o anni? Lo è un pappagallo che parla? O un delfino che gioca? Questi quesiti, così concreti e così urgenti, rimangono ancora irrisolti ed evidenziano un problema generale che ha implicazioni esistenziali e filosofiche: nonostante gli straordinari successi delle neuroscienze, non abbiamo ancora idea di quale sia l’ingrediente misterioso che fa la differenza tra la materia incosciente e quella cosciente.
A questa domanda fondamentale, Marcello Massimini e Giulio Tononi hanno cercato di dare una risposta teorica e pratica insieme. Il principio che li guida, la teoria dell’Informazione Integrata, suggerisce che dentro il cervello c’è davvero qualcosa di speciale, qualcosa che è molto difficile trovare altrove. Ma per dimostrarlo ci vuole una misura empirica. Anche questa teoria, come quella eliocentrica di Copernico, ha bisogno del suo cannocchiale: una sonda magnetica che invia impulsi nel cervello per misurarne la complessità.
Forti di questi strumenti, i due scienziati ci guidano in un’appassionante esplorazione alla ricerca dei confini della coscienza – tra cervello e cervelletto, veglia e sonno, anestesia, sogno, coma, super-calcolatori, polipi, delfini e molto altro. Se la teoria è corretta e le misure non la falsificheranno, questa spedizione potrebbe avere un esito stupefacente: una seconda rivoluzione copernicana che ci riporterà, naturalmente, al centro dell’universo.

Genres: Science, Neuroscience, Nonfiction, Philosophy

208 pages, Paperback

First published August 1, 2013

Reviews

[Arno Klein · Rating 5 out of 5]

This may be the only book I’ve read on the topic of consciousness that introduces a theory about (subjective experience, or the phenomenology of) consciousness from first principles, develops an operational definition for quantifying degrees of consciousness, and outlines years of experiments by the authors themselves.  I really enjoyed this book. 

While I appreciate the symmetrical structure of the book, there are problems with the writing (or translation).  There are numerous typographical errors throughout the book, a glaring inconsistency in delivery, with portions that are delightfully accessible intermixed with portions burdened with unnecessary jargon (particularly for neuroanatomical descriptions), and significant repetition.  For example, I felt that Chapter 4, while it raises interesting questions, could have easily been incorporated in relevant sections of other chapters. Some readers may find the transitions from the brain to theory (Chapter 5) and back somewhat jarring, but I thought the authors provide an appropriate philosophical, social, and clinical context to motivate the development of a meaningful and reliable measure of consciousness that can be generally applied, before revisiting specific brain-related questions.

Chapter 2 introduces the philosophical concept of zombies from David Chalmers’s work, a few variants of “digital zombies”, and “zombies within our skulls”. The cerebellum and basal ganglia are brought up as examples of zombies in our skulls in Chapter 2, but are revisited in further depth in Chapter 6 (see below).  Chapter 3 talks about anesthesia, comas, and minimally conscious states between locked-in syndrome and unresponsive wakefulness syndrome, as well as unsatisfactory attempts to assess consciousness by (1) sensory inputs and motor outputs, and (2) sensory inputs and neural outputs (fMRI, EEG).  Counters to the former include examples of consciousness without sensory inputs or motor outputs. Counters to the latter include false negatives that can arise from many sources, including aphasia, poor attention, damage, mood, confusion, and motion artifacts, as well as awareness with no subject or content (existence without boundary, consciousness as pure presence,...).

Chapter 3 examines unsuccessful attempts at measuring consciousness: (1) global activity levels (counters: sleep, seizures), (2) activity in specific regions of the brain (counters: self-reporting, frontal lesions -- see quote below), and (3) synchronous activity (counters: NREM sleep, anesthesia, generalized seizures, individual exceptions in patients with minimally conscious states).

Chapter 5 introduces the principle of Integrated Information Theory (IIT). The authors contrast the responses of a brain and a photodiode to light and darkness, by highlighting the potential of the brain to differentiate between vastly more states.  Later, they contrast the brain’s integration of information as underlying the unitary experience vs. the independence of individual photodiodes in a camera.  While this characterization is useful for later chapters when contrasting different neural structures, it struck me as a bit of a straw person argument. What if the array of photodiodes were connected to an integrated circuit?  And what if that system were not only integrated, but could differentiate between stimuli, for example if it were to have object/face recognition built in?

The principle of IIT above is broken down into two principles with their corresponding postulates:

“Conscious experience is rich in information.  Thus, the physical substrate of consciousness must be highly differentiated -- that is, it must be able to generate a vast repertoire of states.” (page 66)

“Conscious experience is integrated thus the physical substrate of consciousness must constitute a single, integrated entity.” (page 71)

The authors introduce the quantity phi to define what an “entity” is, which they contend does not already have a satisfactory definition in physics (I believe there have been reasonable attempts in social network analysis to quantify something analogous: clusters of related vertices).  It is expressed in bits, but is different than the information in Shannon’s classic theory of communication.

In Chapter 6, the authors discuss sensory and motor systems, and brainstem activating systems.  As mentioned above, they also revisit the cerebellum and the basal ganglia as examples of “zombies in the skull” (in contrast to the thalamocortical system). They contrast IIT with Global Workspace Theory.

Chapter 7 summarizes years of the authors’ experiments that test integrated information theory on an impressive array of conditions.  They clarify that one has to be careful not to mistake high measures of information with randomness, and not to mistake high synchrony situations for integration if there are parallel drivers that underlie that synchrony.

They list practical rules in assessing integrated information:
----
Rule 1: Observing is not enough, one needs to perturb and detect cause-effect relationships.
Rule 2: Detecting widespread responses is not enough; one needs to assess their differentiation.”
Rule 3: Responses must be recorded on an adequate time scale.
Rule 4: The measurement must bypass sensory inputs and motor outputs.
"A summary of these four guidelines can be formulated as:
Evaluating a brain’s capacity to integrate information requires direct perturbations of cortical neurons (bypass input and output chains) to assess the spatial extent of the evoked response (integration) and its differentiation (information content) on a sub-second time scale (time constant of consciousness).” (pages 100-101)
“Of course, characterizing this echo is still a far cry from calculating phi, which would require perturbing the brain in all possible ways across all possible bipartitions [147, 149].” (page 102)
----

So they introduce their “perturbational complexity index” (PCI) [200]. ”The procedure involved ‘zapping’ the cortex and ‘zipping’ its responses.” (page 120)

The authors even boldly suggest that in our future we will have input and output prostheses to bypass sensory and motor inputs and outputs (pages 134-135).  They also extended discussion beyond structural connections to physiological changes. They go on to describe experiments where they replace transcranial magnetic stimulation with single-pulse intracortical electrical stimulation, and replace EEG with local field potential recordings.

In Chapter 8, the author raises the question as to what other animate and inanimate entities have subjective experiences.  They make references to the attitudes of different philosophers (Thomas Nagel, Descartes, Montaigne, and Peter Singer) about the capacity of non-human animals, and go into some detail about two current approaches, studying the complexity of behavior and the size of the brain.  As with other authors, they believe in graded levels of consciousness: “According to IIT, phi can be graded.” (page 161)

Given their view on animal consciousness and graded levels of consciousness, I was surprised by their stance on embodied vs. simulated consciousness.  While they do not believe that a brain needs a body for subjective experience (contrary to Alva Noe’s book Out of Our Heads, although later they do discuss the importance of experience during development), they are very much against the possibility of software simulation-based consciousness:

----
“Consciousness is not produced through interaction with the external world here and now, it exists in the brain!  If that brain were to be disconnected from its nerves, extracted from the cranium and kept alive in a bath of oxygen and sugar, the dream would continue, rich and bizarre as before, unpredictable as always.  Just as if nothing untoward had happened.” (page 59)

“What about a software that simulates in detail not just our behavior, but even the biophysics of neurons, synapses, and so on of the relevant portion of the human brain, such as the one imagined in Chapter 3?  Functionalism would hold that it would be absolutely conscious, as in this case all the relevant functional roles within our brain, not just input-output relationships would have been replicated faithfully. According to IIT, however, this would not be justified, for the simple reason that the brain is real, but a simulation of the brain is virtual.  Simulating a black hole, will not bend time and space. For the theory, consciousness is a fundamental property of physical systems, one that requires having real cause-effect power intrinsically.” (pages 158-159)
----

There are beautiful passages in this book, particularly in the existential portions at the very beginning in reflecting upon seeing the Earth from the moon, and at the very end:

“Time will tell if another scientific revolution will return us to the center of our universe, naturally and on our merits.  For now, like the astronaut who saw the earth lost in an icy space, the student is caught by a sense of wonder and deep affection.  He would like to shield the weakest flames, the ones that struggle to awake and those that are fading, and decides that the best thing he can do is to explore, integrate and share, to understand the world, and let it exist a bit more.” (page 175)

[Aaron · Rating 2 out of 5]

The author has some ideas that have interesting mathematical significance. Unfortunately, any claims by the author that this work has some rigorous mathematical formalization is presently sorely lacking — at least from anything I have found in published literature. It is interesting to know that one may develop useful clinical tools with this theory as an approximate foundation. However, insofar as a falsifiable theory that produces testable hypotheses, I am not yet convinced this theory predicts heretofore unobserved behaviour — surely necessary of any good theory — nor that it even fully satisfies the broad range of existing behaviours.
The first half of this book might be worth reading if you want to learn about an interesting graph cutting problem.

[Richard · Rating 4 out of 5]

This is the latest in a series of books I have been reading about consciousness. The answers to the questions of what consciousness is, how to define it, where it resides in the brain, and which animals and plants might have it vary greatly from author to author.

In Sizing Up Consciousness, Marcello Massimini and Giulio Tononi focus on the Integrated information theory (IIT and referred to by the symbol Phi). At the core of their theory is the idea that consciousness is determined by its causal properties. Focusing on cause and effect they work to trace how different parts of the brain are activated to move from an experience to a thought or action.

I admit that much of this book was beyond me (and the very small type face definitely did not help). However, I did get a general sense of their theory and I learned quite a bit more about how the brain processes information.

One of the problems created by modern medicine is that we can now keep people alive through medical devices who would have died in years past. This creates the problem of trying to access which people are “brain dead”, which are “locked in” - fully conscious but unable to communicate - and which patients are most likely to come out of a comatose state. The authors do an excellent job of showing how hard it is to decide when consciousness is active and when it is not.

Of course, there is a lot more going on in our brains that what we are conscious of. Nearly all of the functions necessary to keep us alive are not ones that we have conscious control of. This also leads to the question of what happens to our consciousness when we sleep - where have we gone - and where is our conscious mind when our brain is busy dreaming.

The more I read about consciousness the more I realize how complex a thing this is. How do you study something if you cannot define or measure it?

I think that Massimini and Tononi have written an interesting book about their theories and research. While I am glad that I read it, I feel that the presentation was beyond my casual knowledge.

[Ahmad · Rating 3 out of 5]

The theory does one thing well: it provides an objective marker of wakefulness (or what the authors refer to as consciousness). Electrodes are placed on subjects in different states of consciousness (wakefulness, sleep, coma, anesthesia, vegetative state), and TMS is used to perturb a brain area while EEG records how quickly and complexly the activity propagates. This yields the Perturbational Complexity Index (PCI), which correlates well with clinical scores of consciousness. It’s a solid physiological measure of wakefulness—arguably the only useful product of Integrated Information Theory (IIT).

But PCI has nothing to do with qualia. It does not explain why or how experience arises, nor does it address the hard problem of consciousness. At best, it assigns a refined, neurophysiological label to a patient’s level of responsiveness.

That’s not a problem—unless you repeatedly imply and talk about it as if it does.

[Dolf van der Haven · Rating 4 out of 5]

I have for a long time been fascinated by consciousness, both from a spiritual, philosophical and physical perspective. This book was a great introduction into a different approach to unravel the mystery of consciousness, where the authors posit a mathematical-logical model that states that any system supporting consciousness should be able to integrate information, hence the name Integrated Information Theory (IIT).
The book has a wide range of perspectives, ranging from the obvious to the highly complex. It makes a good case for IIT, but, for lack of a definition of consciousness, it remains unclear if their theory actually measures consciousness or just brain functions.
It is a promising route, though, if only to support patients with brain damage, locked-in syndrome or other illnesses reducing their objective consciousness.

[David · Rating 5 out of 5]

Concise, illuminating and unpretentious, this naturalistic theory of consciousness begins with a brief history of the philosophy of mind, smoothly transitions into a explanation of why it is "quality not quantity" of neural networks which results in consciousness, before taking us to the beside of coma patients where IIT has impressive clinical applications, and then back into philosophy of mind and the anthropocentric beliefs of the authors themselves. Lays open the possibility of artificial consciousness. These brilliant neuroscientists render a deep and compelling theory easily comprehensible to layman readers - well worth a read.

[Youssra Belhaj · Rating 5 out of 5]

Assieme sconvolgente ed illuminante.

[Umar Bajwa · Rating 5 out of 5]

Absolutely fascinating, a lot to delve into, mostly charged my inspiration for my first academic paper

[Хамза Пател · Rating 1 out of 5]

Trying to understand why, say, one's experience isn't in line with the cerebellum by contriving such a measurement is like trying to understand biology by contriving a measurement of 'vitality'.

[Makomai · Rating 4 out of 5]

Sconsigliato a lettori facilmente impressionabili*

Questo libro ha il grandissimo merito di proporre una teoria coerente della coscienza evitando ogni forma di dualismo: non esiste nulla di più grande del nostro cervello.
A parte alcune riserve** (probabilmente solo mie), ho trovato tutto assolutamente avvincente e convincente.

*I capitoli introduttivi su coma, stato vegetativo, anestesia e coscienza possono togliere il sonno

**1. Non sono del tutto convinto che sia possibile ottenere una vera e propria “misurazione” della coscienza, il che porrebbe inoltre il problema della soglia sotto la quale la si possa ritenere assente; sostenere d’altra parte che si tratta di una gamma non discreta equivale a ritenere che anche il singolo fotodiodo ha una certa misura di coscienza. 2. Mi sembra che vi sia una certa circolarità, insita nella definizione di coscienza adottata - dalla quale manca inoltre ogni riferimento alla continuità (a meno che non si ritenga che essa sia un portato di complessità e unità); personalmente, ritengo che non abbia molto senso parlare di coscienza se non la si riferisce al sé, ossia ad un’identità nel tempo che per molti aspetti è una nostra invenzione.

[Giulia · Rating 5 out of 5]

Poche pagine da leggere tutte d'un fiato; semplici parole, che giustapposte segnano i passi di un cammino che conduce all'enigma che da sempre aggroviglia le menti dei maggiori filosofi e scienziati: la coscienza. Che si parta dal cogito o dalla teoria dell'integrazione, nulla si toglie alla grandezza di ciò che risiede nel nostro cervello: noi soggetti siamo quel nulla di più grande che domina l'universo, concentrati in un infinitesimo spazio illuminiamo il mondo, ma d'altra parte la storia stessa dell'universo ha origine da un punto di infinita densità.
Massimino e Tononi con questo "saggio" si propongono di offrire un nuovo punto di vista su ciò che accade nel nostro cervello, portando a sostegno della propria tesi esperimenti e conferme senza mai però dimenticare come ogni teoria scientifica non possa mai essere verificata ma solo falsificata; ed é proprio grazie a questa umiltà di pensiero che il lettore non si sente sopraffatto da una serie di dati e tesi, ma al contrario riesce a fare proprio ogni singolo passaggio logico, tanto da essere quasi sconcertato dalla potenza di ciò che ha tra le mani