The Churchlands Consciousness and Naturalised Epistemology

The Churchland's Consciousness and Naturalised Epistemology
Patricia and Paul Churchland's most recent books 'Touching a Nerve' and 'Plato's Camera' respectively are interesting attempts extend their ideas on how neuroscience can expand our understanding of traditional philosophical problems. Patricia's book is written in a more confessional mode than Paul's as she intermingles her scientific evidence with stories from her own life and how her science relates to her experiences. Also her book uses a lot more neuroscientific and psychological data than Paul's does. She brings together interesting evidence on both the nature of morality and in particular on the nature of consciousness.
Paul's book is more theoretical and aims to show how connectionist models, and the more neurally realistic Hebbian learning, can be used to solve traditional problems in epistemology. Paul's book can be read as a neurally focused version of Quine's attempt to Naturalize Epistemology. I have long called for our Naturalized Epistemology to be updated in light of new scientific evidence. Since Quine's seminal paper 'Epistemology Naturalized' was written over fifty five years ago most of the philosophical attention has been on whether Quine's conception of epistemology can actually deal with traditional epistemological concerns. So we have had challenges from people like Stroud that Quine's project reduces to scepticism, and we have had claims that Quine's Naturalized Epistemology cannot deal with normative questions. This back and forth has been very instructive and important; however, I think that it some philosophers have become so obsessed with replying to non-naturalist critics that they haven't developed an adequate Naturalized Epistemology.
As an undergraduate philosophy student whenever the topic of Naturalized Epistemology came up Quine would be mentioned, nobody seemed to care that people like Chomsky had radically different views of how people went from stimulus to science. Few cared that Chomsky dismissed Quine's epistemology as unscientific behaviourist nonsense. Likewise at the same time as Chomsky was busy criticising Quine, the developmental psychologist Piaget was sketching his own theory of Genetic Epistemology. While there were some conferences where people like Piaget, Fodor and Putnam thrashed these issues out, typically in philosophy circles when Naturalized Epistemology was mentioned it was Quine's project which was thought of.
In my own PhD I tried sort out the debate between Chomsky and Quine on the nature of epistemology and naturalism. I also tried to update the debate in light of current research in both cognitive science and developmental psychology. However I didn't deal with neuroscientific data. In this sense I find Paul's book a breath of fresh air. He brings the project of naturalising epistemology much closer to reality by trying to make it neurologically realistic. When philosophers are debating whether epistemology can be naturalized it is Paul Churchland's project not Quine's (no disrespect to Quine is intended; I am merely acknowledging that the science has evolved since his time) that they should be engaging with.
I think that both Patricia and Paul's books should be read together, taken as a whole they offer a detailed introduction to Neurophilosophy and dispel some of the caricatures of the discipline that are offered in the literature. Nonetheless there are aspects of both books that I disagree with to varying degrees.
I think that Patricia's explanation of consciousness is excellent and one of the most compelling that I have heard from anyone on the topic. She notes the simple point that the most basic experience we have of being consciousness and not conscious is the difference between being asleep and being awake. So the study of the activation of the brain during sleep will obviously be an invaluable tool to study consciousness. So we can study and contrast the brain states responsible for being awake, dreaming (REM and non REM), and being in a deep sleep. Sleep is evolutionarily ancient. Even fruit flies sleep, (see Robert Greenspan), also caffeine keeps fruit flies awake and they respond to anaesthetics in the same way as humans do. (Touching a Nerve p. 227).Churchland notes that a good way to study sleep is by studying sleep disorders.
She begins by studying sleep walking disorders. She notes that when sleepwalking our perceptual systems are working well enough to be able to perceptually guide our way around the world, we can open doors, we can find our way to places we know (our friends house etc), but our perception is not normal. We may not see or recognise people on front of us; we may give no indication of hearing what somebody says to us. So while we do have some sort of perceptual abilities it is different from the perceptual abilities of our conscious selves. We also typically do not remember our behaviour while sleep walking. We are also not aware of the bizarreness of our behaviour (e.g. walking down the road naked), while such behaviour would be embarrassing if done while conscious. It would be inappropriate to claim that people who sleep walk are unconscious; rather it makes more sense to say that they are experiencing an alternative form of consciousness. Studying the neuroscience of this alternative form of consciousness will help us to understand the nature of standard forms of consciousness.
One hypothesis is that sleepwalking is a form of active dreaming in which the normal paralysis which occurs during dreams is switched off. Patricia Churchland denies that this is the case. To do this she discusses some contemporary neurological research into the nature of dreams. She discusses how REM does not occur during deep sleep. People who are woken up during REM typically report having dreams. However this fact should not be interpreted as indicating that dreams do not occur in Non-rem States, or in deep sleep states. People who are woken up in deep sleep still sometimes report that they have had dreams. Though their dreams are not typically reported to be as visual during deep sleep. She reports that brain scans of people sleep walking show that the brain activity which occurs during sleeping does not occur when sleepwalking. So she concludes that sleepwalking is not the same thing as dreaming. She also notes the strange fact that the brain activity which occurs during REM are closer to those a person has when awake than a person has when in a deep sleep. One difference between the brain activity of a person who is dreaming and the brain activity of a person who is awake is that the awake person's prefrontal cortex is more active. She correctly notes that this brief discussion of consciousness shows us that an important way to understand the nature of consciousness is to study the neuroscientific difference between people who are awake and people who are unconscious i.e. in a deep sleep. 
She agrees with Dennett that there is no single location in the brain where consciousness occurs, and argues that there are particular structures in the brain (along with the looping links connecting those structures) which are necessary for the maintence of consciousness. She maintains that it is important to distinguish between two things:
(1) The structures that support being conscious of ANYTHING AT ALL.
(2) The structures that support being conscious of THIS AND THAT (the contents of consciousness).
She discusses the work of Nicholas Schiff a neurologist who studies disorders of consciousness. His research has lead him to the central thalamus and its ingoing and outgoing pathways as giving us the capacity to respond to the world. She argues that there are looping neurons from these upper layers of cortex projecting right back to the ribbon in the central thalamus. (ibid p. 235) The looping back allows for maintaining an especially potent but transient connection for a chunk of time.
She argues that some regions of the thalamus that connect to, for example, the visual area do so in a domain specific manner. So the retina is connected LGN in the thalamus, however the LGN projects only to the visual cortex area V1-not to everywhere, not even to everywhere in the visual cortex (ibid p. 235). This is typical of other sensory modalities. She calls this a system by system development of a specific signal. She notes that things are different in the central thalamus, the pattern of the central thalamus suggests a different set of functions: be awake and alert or down regulate and-doze. She makes the point that distinction of functions in the thalamus corresponds with the distinction between being aware of something in particular and being aware of anything.
She goes on to say that there are other aspects of the central thalamus which separates it from other cortico-thalamic systems such as the style of the neuronal activity. The central thalamus has unique connectivity and unique behaviour. Think of this interms of the awake/sleep cycle. During awake and dreaming states the neurons fire in bursts at an unusually high rate (800 to a 1000 times per second). This bursting pattern is not displayed during dreamless sleep. So this strongly indicates that the central thalamus plays a big role in making people conscious.
The ribbon of neurons that is the central thalamus is controlled by the brain cell and regulates the cortical neurons to ready themselves from consciousness. Put succinctly BRAINSTEM + CENTRAL THALAMUS + CORTEX is the support structure for consciousness.
She importantly describes what happens when the Central Thalamus is damaged:
(1) If a lesion occurs on one side of the Central Thalamus people tend not to be conscious of the affected side. If both sides are affected then the person is in a coma.
To be aware of something, say a dog barking one needs to have the brainstem, central thalamus, and upper layer of cortex in its on state. Central thalamic neurons must be firing in bursts that ride the lower frequency brain wave of 40hertz. In addition, the specific areas of the thalamus (for sound and sight, respectively) must be talking to the proprietary areas of the cortex. She asks how we manage to be conscious of a complex scene when a lot of the consciousness must be backed up by unconscious processes. She relies on Bernard Baars Global Workspace Theory to explain these facts.
Baars argued that the important way to begin was by listing the significant psychological properties and capacities associated with being aware of SPECIFIC events.
(1) Our conscious experiences are highly integrated and so rely on a plethora of background unconscious work. (Think of being conscious of what a person says to you; a massive amount of unconscious processing of syntax, phonology etc takes place behind the scenes to determine what you experience).
(2) Your background information about your dog is available to help you decide what to do in a novel situation. So your background information must interact with your perceptual experiences.
(3) Consciousness has a limited capacity. You cannot follow two conversations at once.
(4) Novelty in a situation calls for conscious attention.
(5) Information that is conscious can be accessed by many other brain functions.
Churchland notes that these five proposals together provided an interesting framework to try and study consciousness empirically. This Global Workspace Theory to indicated the importance of a rich integration of information from different sources to yield conscious experience. The idea is that consciousness is a consumer of information from a variety of different sources. This leads us to another question: what is the difference in the brain for non-conscious processing and conscious processing?
One way to test what is going on in the brain when a process is non-conscious instead of conscious, is MASKING TESTS. If the word DOG is flashed to a screen followed by a delay of about 500milli seconds and then xxxx is flashed you will consciously see first DOG then xxxx. However if DOG is followed immediately by xxxx you will not see DOG only xxxx. Experimenters who want to get experimental traction on the difference between conscious and non-conscious processes scan the brains of people who have DOG masked and those who do not. What results have they found?
When the visual signal DOG is masked (not consciously perceived), only early visual areas (in the back of the brain) show activation. By contrast, when the visual signal is consciously seen, the posterior activity has spread to more frontal regions, including pariential, temporal and prefrontal areas. Dehaene and Changeux refer to this as GLOBAL IGNITION. This GLOBAL IGNITION theory is of spreading activation from the posterior to anterior regions nicely confirms Baars GLOBAL WORKSPACE theory which states that conscious perception involves global connectivity in the brain.
Churchland speaks of RICH CLUB NEURONS as being central to making the wiring of the brain cost effective and ensuring that communication between the brain is efficient. An important discovery of neuroscience was that the connectivity underlying global ignition was that all mammals have a SMALL WORLD organisation. NOT EVERY NEURON IS CONNECTED TO EVERY OTHER NEURON. However any given neuron is only a few connections away from other neurons. Some neurons are extremely well connected we call these rich club neurons.
A signal may follow this sequence: from a Locally Connected Neuron to a Feeder Neuron to a Rich Club Neuron to a different Feeder Neuron and then on to a new Local Neuron. A Small World Organization with hubs is more efficient than having every neuron connected to every other neuron. But how does the Rich Club framework help us connect with the Central Thalamus? Well consciousness of sensory signals involves linking spatially quiet separate regions of the brain. As attention shifts, the prevailing linkage weakens and other neuronal pools now take their turn in making strong links. So the burst of spikes displayed by the central thalamus are responsible for catching the gist of a lecture, while the local connectivity of neurons within groups provides the context of the gist. Studies of the effect of Anaesthetics on people support the above model. Anaesthetics work by messing up signals between neurons thus blocking global ignition. There are three properties that are very important in the neurobiology of consciousness.
(1) Rich Club Neurons and their ability to make fast connections to other Rich Club Neurons, thereby providing scaffolding for rich integration of information.
(2) Global Ignition for brain events that reach consciousness.
(3) The Central Thalamus with its role in enabling specific contents of awareness during the awake and dreaming states.
So that in nutshell is Patricia's theory of consciousness. As a summary of the state of play of scientific research into consciousness it is excellent. However from a philosophical point of view it leaves a lot of questions unanswered. Philosophers like, for example, Thomas Nagel, David Chalmers and even John Searle would agree with a the scientific evidence provided by Patricia but argue that it leaves the philosophical questions untouched.
Nagel argues that consciousness emerges from brain processes but the subjective experience of consciousness cannot be reduced to the material brain. While John Searle has argued that consciousness is a biological phenomenon, it is an emergent entity hence it is something different from the material basis that it emerged from. David Chalmers notes that explaining the neural correlates of consciousness is what he calls the "Easy Problem" of consciousness, and it leaves the hard problem (explaining how our actual subjective experience emerges these neural correlates) untouched. So the trio of philosophers I have mentioned could justly complain that Patricia is ignoring the philosophical problems that they are concerned with.
In her 'Touching a Nerve' she does briefly criticize Chalmers' views by arguing that his negative claims about whether consciousness can be explained by neuroscience lead to no real explanatory research programmes. She argues that his entirely negative claims are largely irrelevant to the ongoing scientific research. However, it is imperative as philosophers operating within the space of reasons, that we explore the relevance that this scientific data has for philosophical views on the nature of consciousness.
Patricia Churchland like Paul is famous for adopting the approach of eliminative materialism. However she has also said on many occasions that she is not an eliminativist about consciousness. Her eliminativism about folk psychology, is not as it is sometimes presented, a kind of a priori philosophical position which she adopts. She merely argues that just like some of our folk physics terms were eliminated as we learned more in theoretical physics, so some aspects of our folk psychology MAY have to be eliminated as we learn more about neuroscience. So her claim was an empirical one not some kind of a priori claim. When it comes to consciousness she does not expect a theory of consciousness to eliminate our ordinary conscious experiences, however she does expect that it may in some cases radically alter our intuitions about consciousness. In her 1998 paper 'What Should We Expect from a Theory of Consciousness' she correctly notes that reduction does not always mean elimination. So, for example, she notes that nobody thinks that light is not real as a result of Maxwell's equations. She surmises that like things like pain will not cease to be real as a result of future discoveries in neuroscience. She admits that in the past science has shown that, for example Caloric was not real. So science can show that certain theoretic terms pick out substances and properties which are not real, and she admits that future science may show that our folk-psychology of belief desires is a bad theory which does not pick out a real entity in the empirical world. Yet she claims that future science will not explain away pain. This is a very strange position.
In one sense it is a position which a lot of philosophers and the man on the street would accept. The vast majority of people would agree that no matter what we learn about the neuroscience of pain it cannot take away from the basic fact that pain feels intrinsically awful. Based on this interpretation of Patricia it would appear that while she thinks that beliefs and caloric are theoretic entities which are subject to refutation by future science, our experience of pain is a basic fact not a theory and hence it cannot be refuted by future science.
However the above interpretation of Patricia is difficult to maintain in light of other things she says about consciousness. She admits that our views about the nature of conscious experience have been modified as we have learned more about it through neuroscience. She cites discoveries such as Sperry's about the nature of commissurotomy, research into sleep and dreaming by people like Hobson, as well as the discovery of blind sight and argues that they have cast doubt on our intuitive conception of what consciousness amounts to. However she argues that while these experiments show the need to reconceptualise our basic theory of consciousness they do not eliminate our actual conscious experience (Churchland: What We Should Expect from a Theory of Consciousness).
When she speaks of our actual conscious experiences which remain even after we have modified our theory of consciousness then she seems to being a realist about subjective experience in a similar manner to Thomas Nagel. Her view is uncongenial to the claims that Dennett makes in his 'Consciousness Explained'. Now obviously there is nothing wrong with siding with Nagel against Dennett but it is surprising given her naturalist leanings. I think it is worth exploring this issue in more detail as it is a very important philosophical issue.
Nagel argues that there are facts of the matter about subjective experience even though we may not ever be in a position to know precisely what these facts of the matter are. Furthermore he thinks that there are intrinsic facts about our subjective experiences independently of our theorising about these experiences. Patricia Churchland who holds that our consciousness will not be explained away even if our conceptions of it are seems to have landed in a similar position as Nagel. She seems to hold that there is something given in our experiences independent of our theorising about them.