KOGLIST RE: forward:Toward a Scientific Basis for Consciousness

Date: Fri, 22 Apr 1994 15:32:48 +1000 Reply-To: PSYCHE Discussion Forum &lt;PSYCHE-D%NKI.BITNET(a)UGA.CC.UGA.EDU&gt; Sender: PSYCHE Discussion Forum &lt;PSYCHE-D%NKI.BITNET(a)UGA.CC.UGA.EDU&gt; From: Kevin Korb &lt;korb(a)bruce.cs.monash.edu.au&gt; Subject: Toward a Scientific Basis for Consciousness X-To: PSYCHE-D%NKI.BITNET(a)cunyvm.cuny.edu To: Multiple recipients of list PSYCHE-D <PSYCHE-D%NKI.BITNET> Last week (April 12-17, 1994) I attended the following conference in Tucson, AZ: TOWARD A SCIENTIFIC BASIS FOR CONSCIOUSNESS Here I will present a short review of this exciting meeting, along with some opinions on some of the ideas presented. The overall mood at the conference, about 300 attending, was unusually enthusiastic, due to the fact that this was really the first conference of its type, bringing together researchers from many fields to talk about what has been for the major part of this century a taboo subject in scientific circles. Recent advances in a number of neuroscience methods have made possible the objective, independently verifiable observation of a number of phenomena associated with conscious thought. Furthermore, new modeling paradigms associated with distributed processing in neural networks, chaos, and emergent phenomena have resulted in simulations displaying many of the qualities of conscious entities. These forces, and the fact that consciousness is something we all (hopefully!) have and care about, brought us together and will most likely result in the expansion of this and related conferences in the years to come. Attending were researchers from very diverse fields, including biological neuroscientists, computational neuroscientists, philosophers, anesthesiologists, physicists, mathematicians, psychologists, neurosurgeons, 'chaoticians' (a la Malcolm, in Jurassic Park), cognitive scientists, and a fairly large contingent of 'researchers' of the paranormal (ESP, shamanistic rituals, meditation). I was a little disappointed that the organizers did not weed out a few of the more flaky posters and talks, but the diversity was fun. The majority of the posters and all but a couple of the talks were in the non-flaky category, i.e., relating to testable hypotheses and the scientific method currently accepted in the academic world. There were no parallel sessions, and the auditorium (at the U. of A. Medical Center) was packed, making coffee breaks in the small lobby fairly claustrophobic. A table was set up for anyone to display a book or paper they were pushing, for people to look at. I learned about a number of exciting books that I must go out and get ASAP this way, and think this ought to be done at more conferences. Abstracts for all of the talks and posters were provided along with the program, in a nice binder. It was announced that the complete papers will be published in a book, but I will not hold my breath, as the presenters are not even required to submit them for a month or so. A fairly complete list of those attending was also provided by the end of the conference. There were a number of field trips arranged that were too pricey for my taste. I brought my bike and arranged my own field trips for free. For official info about the conference or proceedings, dont ask me, ask the organizers: Alfred Kaszniak kaszniak(a)ccit.arizona.edu Stuart Hameroff srh(a)ccit.arizona.edu Jim Laukes jlaukes(a)ccit.arizona.edu Okay, here are some observations and opinions about the talks (and posters). A lot of folks gave whole talks on or at least paid lip service to the notion of quantum mechanics having something fundamental to do with consciousness. David Chalmers, in his excellent wrap-up put my opinion of this issue cleverly: "Consciousness is mysterious. Quantum mechanics is mysterious. By the 'Law of Minimization of Mystery', if you find two mysteries, maybe they are the same." I feel like QM may be a useful metaphor for some aspects of consciousness, but we know _way_ too little about either to apply QM literally as a basis for consciousness. Let's be careful not to mix up our metaphors and theories. Relating to this issue was something I had not heard about before (surprising, considering I was a brain biochemist as a grad student) and was quite interesting, was the possibility that microtubules could be computational or information-transmitting elements. Microtubules (MTs) are the major cytoskeletal elements of neurons, and basically all cells. Several talks (somewhat redundantly) described how tubulin, the monomer protein of which MTs are made, can switch between its two conformational states based on the conformational state of neighboring monomers. Anyone who knows jack about cellular autonoma will recognize certain similarities here, and they actually did simulations to show waves of conformational change propagating along the MTs. (Or along organized clusters of water molecules next to or inside the MTs.) As far as I could tell, this was all rampant speculation, yet to be borne out by experimental studies on real MTs, but had a number of testable predictions. Hameroff mentioned that the lowly single-celled paramecium has quite a complex repertoire of behavior (not quite on the level with human consciousness, I would argue!) and yet no nervous system. He proposes that its complex web of MTs may be its 'nervous system'. Being fairly well-read in the neural nets literature, I am struck by how much has been accomplished with models using fairly simplistic 'neurons', and certainly nothing as sophisticated as _sub_cellular nervous systems within them. This leads me to believe that, whether MT processors exist or not, we dont need to invoke such theories to explain consciousness. I must not slight the boatloads of research that has shown that rearrangement of MTs and other cytoskeletal elements is important in learning and memory, but changing the cell's shape is a different issue than what these folks were proposing. Just about every single speaker had a quote from or mentioned the work of William James, who seems to have figured all this out around the turn of the century. I am _not_ well read in psychology, but am inspired to track down his writings, considering how progressive many of his ideas on consciousness were. For instance, he was well aware of the associational nature of all concepts, i.e., nothing has any meaning except in relation to other things. I enjoy taking this concept all the way down to the neural level, where the associations are between activated ensembles and such. I was very pleased to learn of the substantial work of Eric Harth on the importance of feedback connections in the brain in perception. These massive tracts, sometimes bigger than the feed-forward ones, are far too often ignored by neuroscientists, leading them to believe that the brain is stimulus-driven. Well, of course it is, but I feel (and Harth backed my feelings) that most of what we see, feel, hear, etc. emanates from our brain, the environment merely getting the thoughts started. One of my heroes, Walter Freeman, described how a stimulus makes the brain (the smell-centers, specifically) go from a chaotic state to a state of aperiodic oscillation, a basin of attraction probably corresponding to the conscious feeling of recognizing the odor. I bet the feedback connections play a major role in sending the system into its basins. (Perhaps he has knows this already, I am not sure.) I was disappointed that several other prominent Thinkers on Thinking did not attend the conference--were they invited? Notably, Douglas Hofstadter, Daniel Dennett, John Searle, Ray Jackendoff, Stephen LaBerge, the Churchlands, Robert Ornstein, Francis Crick, Geoffrey Edelman, Martha Farah, and Marvin Minsky, come to my mind. Perhaps next year! The conference was mostly synthesis, proposals and speculations, so no great answers about how consciousness happens yet. The best real data presented, IMHO, was from Bruce McNaughton, who records from up to 150 neurons at a time in the hippocampus of a freely behaving rat for weeks at a time. Beside the fact that this is an extremely impressive technical feat (I know because I am trying to do very similar things with cultured neurons, that dont run around all night), it was a real window into the thought processes (consciousness?) of a rat. He recorded the neurons' activity while the rat was foraging around a box, and during naps before and after the foraging. He found 'place cells' that fire only when the rat is in a specific part of the box. He was able to use the place cell 'map' to accurately predict the rat's trajectory, based on the neural signals. He also showed the effect of subsequent learning on previously established maps. But even cooler was the observation that neurons that had correlated firing during the foraging and not before (implying that they are learning preferences to nearby places), also were highly correlated during the post-forage nap. Thus, the rat may have been dreaming of snippets of its box experience, reinforcing or consolidating the important associations. Neat stuff. I look toward functional MRI to come up with similar experiments on humans in the near future. Keep your eye on this amazing new technology: I predict that it will oust PET studies in a year or two, due to its far greater spatial and temporal resolution, and the fact that it does not require the poor subject to dose up on radioactivity. An excellent conference overall, with many speakers referring to other's talks in their talks. Subscribe to Psyche-D if you are into this type of thing and want to chat with like minded individuals who obviously have way more free time than I do. Steve Potter, Ph.D. Division of Biology 156-29 California Institute of Technology Pasadena, CA 91125 spotter(a)druggist.gg.caltech.edu