Stanislas Dehaene, a French neuroscientist at the French College who has worked with neuroscience pioneer JeanPierre Changeux (author of "Neuronal Man" and "The Physiology of Truth"),
wrote the pop-science best-seller "The Number Sense" (1997) and edited the influential compilation "The Cognitive Neuroscience of Consciousness" (2001).
His "global neuronal workspace"
(originally proposed in "A Neuronal Model of a Global Workspace in Effortful Cognitive Tasks", 1998)
is an elegant compendium of contemporary
neuroscientistic findings, from Donald Hebb's cell assemblies to
JeanPierre Changeux's and Gerald Edelman's neural Darwinism
via Bernard Baars' global workspace and Antonio Damasio's convergence zones.
This new book is admirable because it is not technical (at all) but it talks about the technicalities of research on consciousness. There are many books that discuss consciousness, and it is one of the easiest things to do (virtually any human being equipped with consciousness and language can have a valid opinion on it); but very few do it from an experimental perspective. The only warning is that the book is full of "recent studies", "recent experiments", etc: everything recent in science is not... science. Science takes years and sometimes decades to prove the validity of a theory. Anything "recent" is just a theory. The experiment needs to be reproduced by many other labs, and survive the critique of those who have done different experiments with different results.
Nonetheless, Dehaene belongs to the generation of neuroscientists who think that we can have direct access to the nature of consciousness, thanks to new machines and new techniques. In fact, it is not difficult to manipulate your consciousness and make you see or not see things that are in front of your eyes. The "subliminal" messages are precisely one practical application of this fact. Dehaene explains how he can make words, pictures and even entire stories disappear from your consciousness.
One area of research is therefore what happens in the transition from unconscious to conscious, and viceversa. What changes in my brain so that i become conscious of typing on a keyboard when a few seconds ago i was typing mechanically? And what changes in my brain when i start focusing again on the screen and stop being conscious of what my fingers are doing?
Dehaene begins by introducing some technical terms, as "consciousness" is not quite a scientific quantity that a scientist can measure. "Conscious access" is the one thing we are conscious out of an infinite repertoire of potentially conscious events. Conscious access is not "selective attention": attention selects which of the many unconscious processes becomes conscious and therefore by definition it operates unconsciously. There is also the generic state of being conscious, which is more properly called "vigilance" or "intransitive consciousness": there are many gradations of being conscious, from sleeping to very focused. Vigilance corresponds to the level of excitement going on in the cortex and in the thalamus. Conscious access is enabled by vigilance and attention. Conscious access is when we have a thought. Attention is what unconsciously selected that thought. Vigilance is what made it possible for us to be conscious.
We are not conscious of most of what our brain does, of most of what our organs perceive and process. Dehaene studies how deep into the brain does the unconscious travel. He found that a lot of "intelligent" processing occurs unconsciously. People even do math while they are unconscious of doing it. A word can travel deep into the cortex and affect what we think and do. The unconscious is not simply the hell of traumas that Freud imagined but a sophisticated computer that performs a lot of operations automatically on all the perceptions of all the organs. Therefore Dehaene's studies lend credibility to the idea that some great insights by scientists are incubated unconsciously. After all, we know that sleep is not just a state of relax but a state of intense processing of memories. The unconscious performs "brute processing" of multiple parallel sensory inputs. Consciousness selects one of these preprocessed pre-thoughts and runs with it. "The unconscious mind proposes while the conscious mind selects". He compares consciousness to the spokesperson of a large organization, where thousands of people are processing huge amounts of data, but the spokesperson gives one simple linear version of what is going on. Consciousness removes all the ambiguities, uncertainties and contradictions from the input data and creates a simple coherent truth.
Dehaene therefore identifies an evolutionary purpose for consciousness: "consciousness selects, amplifies and propagates relevant thoughts". It was selected over millions of years of evolution because it serves a purpose. Dehaene believes that consciousness allows humans to learn over time rather than living only in the moment. This is the weakest part of the book. All he proves is that we think the way we think because our brains work the way they work (long-term memory, consciousness, etc). Looking at the (brief) history of the human species so far it is not clear that consciousness is really useful for survival. In fact, many people (who have no reason to complain about their material situation) end up committing suicide when they exercise the faculty of consciousness too much, whereas others strive to live as long as possible despite living in horrible conditions (war, famine, poverty, etc). And this is particularly true of teenagers, that in some developed countries are more likely to die of suicide than in a car accident. Dehaene also speculates that consciousness may facilitate sharing knowledge with other members of the group, but that is also a weak argument: machines do that all the time but we don't assume that they are conscious. Dehaene instead thinks that, to an extent, our self is shaped by the knowledge that we receive from others.
The most interesting part of the book begins in chapter 4, where Dehaene explains how he looks for the "signatures of consciousness". Several neuroscientists are looking for the "neural correlate of consciousness" but that is not the same thing: a "correlate" is not necessarily a cause. If i push you down the hill, stones will fall with you: they are a correlate of your fall, but not the cause of it. Dehaene is looking for the cause of consciousness, not just for what happens when we are conscious.
In 2005 he and Claire Sergent pinpointed the P3 wave corresponding to massive positive voltage ("Timing of the Brain Events Underlying Access to Consciousness during the Attentional Blink", 2005), first described by the US psychologist Samuel Sutton ("Evoked Potentials Correlates of Stimulus Uncertainty", 1965), as a signature of consciousness. A sign of consciousness is when several regions of the brain fire in a synchronized manner: Dehaene believes that synchrony facilitates the transmission of information. This idea was first advanced by the German neurophysiologist Pascal Fries: neuronal communication is implemented via neuronal synchronization ("A Mechanism for Cognitive Dynamics", 2005). Donald Hebb in 1949 already mentioned the possibility that a network of neurons exciting each other may create a phenomenon of global synchronized activity. Dehaene's "global ignition" theory is then that during conscious experience groups of neurons begin to fire in synchrony and then the synchronized firing extends throughout the cortex and that causes the spike in voltage that is recorded as a P3 wave. There's a tipping point of neuronal activity beyond which a state of global synchronized activity quickly settles in. This point is equivalent to the "phase transition" in physics and to the "bifurcation" in mathematics. Conscious awareness of a stimulus happens about a third of a second after the stimulus has been received, the time that it takes for that "global ignition" to reach a stable state.
Crick and Koch theorized that consciousness arises from synchronized firing of neurons at 40 Hertz (see my reviews of Christof Koch's "The Quest for Consciousness" and "Consciousness"), but Dehaene found that there are also unconscious states that fire like that. Consciousness arises from an "amplification" of this activity that takes place towards the end of the signal processing. The synchronized firing at 40 Hz (and, more generally, of the gamma band) is necessary but not sufficient: it also takes the sudden surge for the brain to yield a conscious event.
Theoretically, it should be able to decode the brain activity corresponding to consciousness. A single neuron can be shown to fire when the brain is presented with an image of a famous building or with the written name of that building, as shown by Itzhak Fried and Rodrigo Quian Quiroga ("Invariant Visual Representation by Single Neurons in the Human Brain", 2005). In other words, scientists will someday be able to "read" what you are thinking.
Neuroscientists can already selectively create or destroy memories, i.e. manipulate what you believe. In 2010 alone the British psychologists Paul Taylor and Martin Eimer created a conscious percept ("The Neural Signature of Phosphene Perception", 2010) and the Chinese psychologist Hakwan Lau destroyed a conscious memory ("Theta-burst Transcranial Magnetic Stimulation", 2010).
Dehaene thinks that consciousness does not originate from a specific region of the brain but it is due to the action of giant neurons whose axons spread throughout the brain, especially the ones in the pre-frontal cortex, as described by the Swiss neuroscientist Stephanie Clarke ("Direct Interhemispheric Visual Input to Human Speech Areas", 1997) and by the Australian neuroscientist Guy Elston ("The Pyramidal Cell in Cognition", 2001). These giant neurons spread information to many regions of the brain, and, when enough of these regions agree that some information is important, they synchronize into a state of "global communication". And that's when the P3 wave appears.
You can potentially think an infinite number of thoughts. Each of these potential thoughts must already be coded in your brain, even those that you will never think. This is made possible by a process of integration that pulls together "information" stored in both neighboring and very distant neurons, and integrated in higher cortical regions similar to Antonio Damasio's "convergence zones" (see my reviews of "The Feeling of What Happens", "Self Comes to Mind", "Descartes' Error", "Looking for Spinoza") with feedback going back to the lower regions in a manner similar to Gerald Edelman's "reentry" (see my reviews of "Neural Darwinism", "Wider than the Sky", "Topobiology", "The Remembered Present", "Bright Air Brilliant Fire", and "A Universe of Consciousness"). In fact, Dehaene's model is not all that different from the model that Giulio Tononi and Gerald Edelman proposed in the same year ("Consciousness and Complexity", 1998). Your cortex has 16 billion neurons and they can be combined in many different ways to yield trillions of potential thoughts. The integration is what turns neural events into consciousness (echoing Giulio Tononi's motto of "Consciousness as Integrated Information", 2008).
It is not enough for consciousness that information be interconnected in the higher regions of the cortex: this interconnection must also ignite a loop leading to a global workspace. The P3 wave does not represent the network of networks, but the escalation of the loop. The coherent and conscious thought is the result of this "ignition".
Furthermore, the giant neurons that organize our conscious life are active all the time. We "think" all the time, even when we are not aware of it. JeanPierre Changeux and Gerald Edelman have explained that the brain is not simply a reflex machine, reacting to stimuli: it is a Darwinian machine, constantly producing "thoughts" that are then selected by the stimuli. Our brain generates thoughts all the time. This random "noise" helps increase the Darwinian variation and the chances that we will have the best possible thought for the next situation that we have to understand.
Dehaene's theory that "consciousness is global information broadcasting within the cortex" is an extension of Bernard Baars' "global workspace" (see my review of "A Cognitive Theory of Consciousness"). The main difference is that Dehaene's "global neuronal workspace" is a hierarchy of neural networks (as explained in his original paper/manifesto written with Changeux, "Hierarchical Neuronal Modeling of Cognitive Functions", 1998).
Dehaene's model admits five types of unconscious, not just one: the preconscious (the enormous number of thoughts that are randomly generated and ready to become conscious), the subliminal (processed stimuli that cannot generate the P3 wave of conscious thoughts and therefore can never become conscious), the disconnected (brain processes that are simply not connected with the cortex in a way that would give them a chance to become conscious, e.g. breathing and heartbeat), the diluted (the multitude of firing patterns that react to stimuli but need to be processed in working space in order to yield conscious thoughts), and the dormant memories that can become conscious again at any time.
The last chapter asks questions such as: are new-born babies conscious? are other animals conscious? and can machines be conscious? His answer is "yes" to all of them. The third one is peculiar. He argues that David Chalmers is wrong in thinking that the subjective feeling of consciousness is the "hard" problem of consciousness. (See my reviews of Chalmers' "The Conscious Mind" and "The Character of Consciousness"). Dehaene's argument is that it is possible to build a machine that has qualia and free will but then his definition of qualia and his definition of free will are almost the exact opposite of what most of us have. His free will is not free will at all: it is the illusion of having free will while performing mechanically a program that comes from genes, life history and neurons. His concept of "autonomous decision making" is that we think we are making decisions autonomously, but that may or may not be the case, and that we are entitled to claim "free will" regardless of whether we are truly in control of our actions or not. In that sense there is no doubt that a machine can do the same: they already do. His concept of qualia (the feelings) totally eludes me. He seems to think that feelings are simply the physical processes associated with them. He doesn't seem to understand the question of where the feeling "comes from". He writes that "The science of consciousness already explains significant chunks of our subjective experience". Really? Where? Who? Certainly not in this book. I haven't seen a single sentence in this book that explains any part of my subjective experience. One is left with the doubt that Dehaene may not be conscious, after all. Suddenly, one gets the feeling (yes, feeling) that this book was written by a robot programmed to study the brain. The robot has done a wonderful job of answering every question that we asked. But the robot is not conscious, so the robot has simply studied the physical structure of a conscious brain and is reporting back, without actually understanding that the question was not about the physical structure but about... consciousness.
TM, ®, Copyright © 2016 Piero Scaruffi