An interview with Patricia Smith Churchland
Written by Thomas Zoëga Ramsøy
Ramsøy: In your earlier bestselling book, “Neurophilosophy”, you presented a significant view of how the study of consciousness should be conducted. As has been said about this book, it “launched a subfield” within consciousness studies. While this book can be said to be more academic in its presentation, your recent book, “Brain-Wise: Studies in Neurophilosophy” seems to focus more on the pedagogical aspects of your approach. What was the rationale for writing this book?
Churchland: Hello Thomas — Thanks for your first question.
Neurophilosophy was my first large-scale attempt to explain to philosophers why – and how – neuroscience is relevant to answering questions about the nature of the mind – or as I prefer to say, the mind/brain. I also wanted to show neuroscientists how to address philosophical skepticism concerning the idea of explaining mental phenomena in neuroscientific terms. Many philosophers wanted to argue about whether to naturalize metaphysics and epistemology; I wanted to do it, not just talk about doing it. So Neurophilosophy was my attempt to do it. This meant that some philosophers conveniently wrote me off as not really a philosopher!
In the 1980’s, work in the philosophy of mind was wholly dominated by those functionalists who believed that details about neurons, neuronal connectivity and the brain more generally, were irrelevant to understanding the nature of the mind. At the time, this seemed to me a ridiculous assumption, and it seems so still. For a range of reasons, mostly sociological, epistemology and metaphysics, allegedly the ‘core’ of philosophy continues to be dominated by what I politely call the “no-brainers”; that is, people who want to avoid knowing anything about the brains in their heads.
I undertook to write Brain-Wise for three reasons: (1) Neurophilosophy needed an update, especially with respect neuroscientific data. There have been many important developments in neuroscience students need to know that there are other games in town besides what is advertised as “conceptual analysis”in the last twenty years, and I wanted to integrate these into the picture. (2) Graduate students and undergraduates need a decent introductory (or introductory+) text that shows how and why neuroscientific data are relevant to long-standing, traditional problems in the philosophy: the nature of consciousness, the self, free will, knowledge, ethics and learning. “No-brainism” is not only boring; it is being left in the dust. (3) Neurophilosophy is sometimes a difficult read, especially in the philosophy sections. I thought I could make the philosophical arguments more accessible, particularly to those students choosing a career in neuroscience.
As far as philosophers were concerned, I also had a larger aim of challenging their standard way of doing business. I wanted to show that much of what academic philosophy continues to honor as “conceptual analysis” is actually hocus-pocus: it advertises itself as mere clarification of concepts, but it secretly winches its way along by theorizing about the nature of things, such as consciousness, knowledge, and free will. When challenged with the facts from neuroscience, it says, “well, conceptual analysis is just about meaning, not facts”. When challenged that the analyses do not correspond to recognizable common meaning, it says “well, this is about what is conceptually necessary, and we explore that with ‘thought experiments’.” But what anyone thinks is conceptually necessary – and the form his thought experiments take – depend crucially on what he believes are the facts of the matter. Conceptual analysis as typically practiced in the last thirty years is by and large a shell game. Clarity is desirable, and theorizing is desirable, but you cannot make progress if you protect your theory from factual criticism by pretending you are only doing “conceptual analysis”. Moreover students need to know that there are other games in town besides what is advertised as “conceptual analysis”.
Ramsøy: So, on the one hand, philosophers seem to have less than optimal knowledge, or are even agnostic, of research in the cognitive neurosciences. On the other hand, scientists may think that philosophical issues are interesting but do not have any direct consequences for their own work. How is Neurophilosophy thought to solve this problem? And do you feel that this approach has had any success?
Churchland: My sense is that philosophers — the young ones anyhow — are beginning to realize that Quine was right, for example about epistemology and metaphysics. The younger generation of philosophers – e.g. Rick Grush (emulators), Thomas Metzinger (self-representation), Steve Quartz (epistemology), Chris Eliasmith (neural computation), Kathleen Akins (color vision), Bill Casebeer (neuroethics) – have taken on projects that are squarely in the interface between philosophy and neuroscience; they have plunged into empirically-based theorizing. And all of these young philosophers collaborate with others in the relevant science. The neuroscientists (e.g. Read Montague, Peter Dayan, Terry Sejnowski, Charles Anderson, and many others) welcome the collaborations. I think this is what philosophy was like for most of its history – until dry and dreary “know-nothing” philosophy became a fad in this century.
Ramsøy: What problems in neuroepistemology do you see as especially puzzling?
Churchland: The nature of knowing in nervous systems must be studied at many levels, from molecules and synapses to the whole organism in its environment. I am particularly puzzled by three major questions: (1) how do neurons code information, and (2) there are many dimensions of neuronal plasticity – how is plasticity orchestrated within the individual neuron, and (3) how do local (neuronal) changes result in global (cognitive and behavioral) changes; i.e. how is plasticity orchestrated across many neurons. None of these questions has been solved, and the more we learn at the level of the synapse, receptor and ion channel, the deeper the coherencing problem becomes.
The problems about the nature of consciousness that interest me most also involve coherencing. For example, what makes for coherence across modalities, as for example when visual and auditory signals are coordinated when we observe someone clapping or speaking, and for coherence within modalities, such that color and motion pretty much stick together when you see a moving car; what are the roles of top-down pathways in developing the content of one’s consciousness; e.g. when one sees a flashed stimulus as the word “devil”. I also want to know whether there are differences in types of coherencing as a function of whether the state is conscious or nonconscious.
Ramsøy: Are there any findings or theories that especially come to mind?
Churchland: Certainly others are attacking this problem, for example by exploring what oscillations or synchrony of firing might achieve. But these are early days, and I have not yet a whole lot of basic slogging, and the concomitant development of new tools, is essentialseen anything that in my opinion brings the problem to its knees. One major obstacle concerns the undeveloped state of micro neuroanatomy, especially in humans. Until we have a clearer understanding of the “back-projecting” pathways, of what cells types exist in cortex, of the nature of the connectivity between cell types and when it emerges in development, of the degree to which pre-frontal structures are hierarchical, and so forth, it will be difficult to know how to ask the right neurophysiological questions.
Structure is always a huge clue to function in biology, and we are currently trying to do neurophysiology in the structural dusk. New tools from molecular biology will undoubtedly appear that will make it possible to identify cells types, for example, and that will help enormously. But a whole lot of basic slogging, and the concomitant development of new tools, is essential. The trouble is, neuroanatomy does not seem very glamorous to many young scientists, and indeed it seldom makes the high spots of the scientific journals. But having structural data is not an option if we want to solve problems such as coherencing. You can theorize until the cows come home, but without the constraints of neuroanatomy, much of the theorizing is just blowing in the wind.
Think of it this way: data space is infinite. Since you do not want to waste time, you want to explore a section of data space that is likely to give you answers to your questions. The best way, in biology, to narrow the search space is to know as much as you can about the structure of the thing that produces the phenomenon. Then you can begin to see how the phenomenon arises from the interaction of the elements. Molecular biology and developmental are spectacular demonstrations of wisdom of this approach.
At the psychological level, one rather remarkable result, discovered by Amir Raz, is that following hypnotic suggestion, the Stroop effect disappears. That is, hypnotized subjects who are told that the strings of letters in colored font are meaningless did not exhibit the usual semantic/color conflict and easily named the color of the print. Crudely put, if you expect the strings to be meaningless, they are seen as meaningless. Which means that the allegedly “obligatory” reading of meaningful words is not, after all, obligatory, but is sensitive on “top down” influences of a certain kind. This result is very surprising because the Stroop effect is exceptionally robust. On the face of it, this result seems to imply a broadly distributed effect on perceptual processing resulting from a high-level cognitive “set”. And indeed, when Raz and colleagues compared fMRI data from experimental and control subjects, they found decreases in conflict (color of print vs a color-word) corresponded to decreased activity in anterior cingulate cortex, believed to play a crucial role in attention. They also found reduction in activity in prestriate areas of visual cortex. I find these results important because they are suggest top-down influences on coherencing.
Ramsøy: Coherence has also been a hot topic in consciousness science during the last, say, ten years. Oscillation hypotheses such as Crick & Koch’s Gamma hypothesis, and more recently Freeman’s notion of neural ‘wave packets’, are often cited as one of the likely mediators of coherence in the brain. From other sources we have instances of breakdown in coherence or even non-regular coherence between areas of the brain, such as in synaesthesia. In brain imaging research we now see a rapidly rising interest in functional and effective coupling (in the so-called dynamic coupling analysis) between brain areas during cognitive or perceptual tasks. In sum, there is a rising interest in coherence and cooperation between areas of the brain. Do you think there are any strong candidates in explaining coherence, and what coherence actually means?
Churchland: I share the hunch that coherence in timing of neural activity is a good place to begin looking for answers regarding phenomenological coherence. I think most people in neuroscience believe that we have only begun to understand the role of time. For example, in Ralph Siegel’s lab at Rutgers University, four different analyses were used to interpret the responsiveness of neurons in superior temporal polysensory area to optical flow. What they discovered was that if you just used mean firing rate (MFR) over 500 ms, you would conclude one thing, but if you binned the spikes at various time bins (between 30 and 500 msec) then sensitivities of some cells, previously hidden on the MFR analysis, clearly emerge. Some cells were sensitive to one aspect of the moving stimuli early in the recording, and another aspect slightly later. Unless we do temporal analyses of data, we are not going to see this.
In a different example, a study of place cells in the hippocampus showed that as the animal came nearer to the cell’s place field in the maze, its spike phase relative to the t theta rhythm gets earlier (J. Huxter and colleagues). This is a fine demonstration one aspect of a stimulus is coded temporally.
I am also eager to see where the Freeman “wave-packet” hypothesis goes. Somehow we need to find the tools to connect what is seen at the macro level with methods such as fMRI with what is seen at the level of single cells.
Ramsøy: While great progress is being made in cognitive neuroscience in understanding the brain and its relation to cognitive processes, it seems that in consciousness science, we are still left with the enigma of ‘how the brain produces experiences’. This notion is also reflected in how consciousness is studied scientifically; we operate with a relatively well developed vocabulary and methodology for studying the brain and cognition, but one might claim that little progress has been made in the scientific study of experiences during the past 100 years or so. In terms of these aspects, does Neurophilosophy propose any future directions for studying and discussing the mind and the brain?
Churchland: Many fundamental questions about nervous systems remain unanswered, such as how neurons code information, and what representation is for nervous systems. There is much we do not understand about brain organization, such as the role of back projections, and the degree to which any subsystem is hierarchically organized. We still do not understand why brains sleep and dream. Although I would like to understand what exactly it is for a brain to be consciously aware of something, such as visual motion, I am patient. I would not be surprised if some fundamental questions in neuroscience will need to be answered before we can get a sure grip on the right sorts of experiments to conduct to make real progress on consciousness. I do not know that this is so, but I suspect it may be so. It is fashionable among some philosophers to assume that if we have not solved the problem by now, then that is telling us something really deep, such as that the problem is not solvable at all. Of course no such thing can be inferred at this stage.
© 2004 P.S. Churchland & T.Z. Ramsøy
Patricia Smith Churchland is the UC President’s Professor of Philosophy at the University of California at San Diego. She has written and co-edited several articles and books, such as “Neurophilosophy: Toward a unified science of the mind-brain”, “Consciousness at the crossroads”, “The computer and the brain”, “Mind-brain continuum”, and “Consciousness and the brain”. “Brain-wise” is her latest book.
For a full list of Churchland’s books, please visit:
Links to texts about or by Patricia S. Churchland
- Can neuroscience tell us anything about consciousness?
- The Hornswoggle problem, article
- Review of Brain-Wise by Human Nature Review
- Reviews/opinions on Brain-Wise at phil-books.com
- Ned Block’s Review of Neurophilosophy or Philoneuroscience
- No more folk tales, interview
- Animal consciousness, radio debate (.mp3)
- Debate with eliminative materialist philosopher Patricia Churchland, articles
- Plutonic: Unifying Concepts of Mind and Brain, interview
- Closer to Truth, debates