Toward a Science of Consciousness

8–13 April 1996
Tucson Convention Center and Music Hall
Tucson, Arizona

An edited version of the daily reports by Keith Sutherland (and on Saturday by Jean Burns), made throughout the conference to JCS-online. Introduced and edited by Anthony Freeman

"Tucson II" was above all else a celebration. The best part of 1,000 delegates from all over the world (though with a heavy Anglo-Saxon bias) and from a wide variety of academic disciplines came together to celebrate the "coming of age" of consciousness studies. This was highlighted by John Searle who recalled the time -- not so long ago -- when those engaged in consciousness research had to disguise their interest in order to retain academic respectability, and he rejoiced not just to be at the conference, but also in the fact that the gathering could be held at all.

A celebration party is a wonderful thing, but it does impose its own limitations. Four dozen plenary speakers, plus over one hundred concurrent talks and several hundred posters, all squeezed into five and a half days, left one feeling breathless and battered. No-one had more than half an hour to speak -- most had less -- and that made for real problems when the material was often technical and the audience largely non-specialist. It was never going to be the occasion for careful analysis of new research, but many delegates reflected that rather fewer speakers, with longer to develop their themes and with time for more extended panel discussion, might have offered greater depth and insight than was in fact possible.

At the practical level, a great debt of gratitude is owed by all participants to Jim Laukes for the smooth administration and his calm and prompt handling of any hitches which did occur. And the ulitimate credit must go to Stuart Hameroff, from the Department of Anesthesiology and his colleagues at the U. of A. and on the organizing committee.

Monday 8 April
The conference started off with a session on the "hard problem" chaired by David Chalmers. It struck me as nothing short of remarkable that there was a clear consensus among the speakers (with one exception) that consciousness could not be explained within our current science.

The first speaker was Michael Lockwood who, in "the enigma of sentience", argued that the scientific view of the brain could not account for qualia. Then Jeffrey Gray gave a resume of his recent BBS article on the neuropsychology of the contents of consciousness. He proposed an experimental programme to try to find out whether functional or structural brain features were responsible for the contents of consciousness, but concluded that neither this (or any other programme he could think of) could tell us why an brain processes should give rise to conscious experience.

The next speakers were Roger Shephard and Piet Hut, who opted to turn the hard problem upside down and propose that brains arise out of consciousness rather than the other way round. The mystery is not our own conscious experience (this is the only thing we know), but the physical world, and the theory dismisses the commonplace observation that consciousness is a product of complexity. What is complex about a simple quale like a pain or the smell or a rose? They then reviewed a variety of psychological evidence from blindsight to jokes in dreaming, before moving on to a review of the sort of physics that would be needed to make sense of conscious experience. This paper has been submitted for the JCS hard problem series.

The only dissenter from the mysterians on the panel was a certain Daniel C. Dennett, who described himself as "like a policeman at Woodstock". In his early career he had liked to view himself as a radical, and it amused him now to find himself defending orthodox reductive materialism against a solid front of mysterians. He argued his familiar case by metaphor and analogy with his customary charm and panache. It was a great performance, but ultimately failed to convince this reviewer.

Then followed around 30 concurrent talks and several hundred posters.

Tuesday 9 April

The morning plenary shifted to the neural correlates of consciousness in which Rodolfo Llinas, JA Hobson, Joe Bogen and Susan Greenfield presented their own very different angles from the viewpoint of brain research. Llinas reviewed the evidence for binding as a result of temporal sychronization (see Thomas Metzinger's chapter in "Conscious Experience") and Hobson outlined his Conscious States Paradigm work on sleep and dreaming as mediated by changes in the brain stem. Joe Bogen, veteran colleague of Roger Sperry and sounding a dead ringer for John Wayne, put forward the thalamic intralaminar nuclei as the best candidate for an area of the brain that was most closely associated with conscious experience. Susan Greenfield took issue with this, arguing that consciousness was more likely to be associated with global brain gestalts than any particular anatomical area. Both Bogen and Greenfield took pains to emphasise that their work only showed correlation and they hadn't the first idea how any brain state could give rise to conscious experience. Greenfield argued that this is one question we shouldn't even ask, as this knowledge could give rise to attempts to manipulate other people's consciousness.

The next plenary was a showdown between a professor of physiology, Colin Blakemore, and a Supreme Court judge, David Hodgson. Blakemore revealed a wealth of evidence showing that our folk-psychological notions of conscious control were an illusion. Unfortunately he spent so much time presenting the evidence that the discussion was a bit thin, concluding that the folk-psychological underpinnings of the law were a thinly-veiled cover-up for an Old Testament desire for vengeance and punishment. David Hodgson made a quiet but passionate plea for scientific and philosophical elites to consider the social and ethical consequences of their theorising. Since the western legal and ethical system was entirely underpinned by notions of free will and volition, theorists had an obligation to consider their evidence very carefully. He concluded that the case against folk psychology was at the moment unproved, but that we needed a new understanding of causality which would find a place for volition and intentionality within our understanding of physics and psychology. He then gave an outline of his own candidate for such a theory.

One of the most exciting panels so far was the machine consciousness session. The moderator, Christof Koch, instructed speakers to keep it short, leaving a delightful 45 minutes for wrangling. It struck me that the arguments from the proponents of machine consciousness were a very thin cup of tea. Dave Rumelhart, one of the pioneers of connectionism, pointed out that cognition in general and emotion in particular had to be seen in an embodied context. He then presented a model of emotion based on the sympathetic/parasympathetic nervous system, but admitted that was all it was -- a model -- and remained sceptically agnostic as to whether any such model might be conscious. Danny Hillis took a much more up-beat approach, arguing that sceptics of machine consciousness were guilty of a lack of imagination. However he didn't explain why it was any more valid to "imagine" that consciousness might result from some as-yet-unconstructed computational system rather than any other slice of the physical universe. My own understanding was that science was to do with evidence and that over-fertile imaginings were the province of people with very different white overalls who worked with padded cells, not brain cells.

Jaron Lanier seemed to win over the audience with a delightful send-up of computationalism. But the humour hid a deeply serious agenda. Lanier has argued in JCS [Vol. 2 (1), pp. 76–81] that the AI research programme has had a damaging impact on our view of ourselves and a "nerdy" effect on our culture. Roger Penrose then finished with a defence of his own theory that human understanding has nothing to do with algorithmic processes.

Then the fun started. Christof felt he should perhaps stir up a good argument, but he needn't have bothered. Dan Dennett was first to the microphone, replying to Lanier's last JCS piece [Vol. 2 (4), pp. 333–44]. Dennett was happy to agree with Lanier's view of his zombie philosophy and challenged us all to clarify what it is that we claimed "we" were over and above our own zimbo mental processes. He also agreed with Lanier's observation that the ascendancy of AI has been correlated with a return to religious fundamentalism, but he said that this was no reason for scientists and philosophers to exercise some sort of self-censorship. Susan Blackmore then launched a passionate attack on another of Lanier's claims. But the most memorable part was an exchange between Penrose and Hillis during which the back and forth of the table microphone was reminiscent of a John McEnroe final on the centre court at Wimbledon.

Wednesday 10 April

Today's session started with a panel on Language and Interspecies Communication. Colin Beer, the moderator, introduced the session with a short overview of some of the controversies in recent ethology. In 1976 Donald Griffin proposed that we should try to communicate with animals in their own tongue, but this approach was not found to be effective and has now largely been replaced with the attempt to teach animals to use human speech and communication systems.

Sue Savage-Rumbaugh began by discussing how Descartes' view of animals as devoid of consciousness had led to the view that there was an innate grammar module in humans but not animals. The communications systems of all other animals was thought to be closed, hard-wired and only "informative" to others of the same species in an unconscious or unintentional sense. But her own work among bonobo apes had led her to the view that apes can spontaneously acquire a human language system if exposed from an early age, and that conscious intentional language usage is clearly within the cognitive competency of apes. One of the most interesting experiments that she outlined was a Piagettian test which showed that bonobos possessed a "theory of mind" which was as good as human 4-year olds. Her bonobos used symbols in novel and context-dependent ways so she concluded that by focusing tightly on syntax and human speech we have come to myopic conclusions on the true nature of language.

Diana Reiss outlined some of the work that was being done with dolphins. Although this species is far removed from humans there is clear evidence of linguistic systems. Although there was a danger of reading too much into anecdotal evidence, her own observations and those from other laboratories on mirror use and play suggested evidence of self-awareness and self-monitoring.

Irene Pepperberg presented some extraordinary evidence from 20 years' work with a grey parrot called Alex. The bird showed a highly-developed grasp of categories such as colour, shape, same/different, bigger/smaller, quantity and even confounded number sets. He was able to apply these categories to novel objects with a consistent accuracy of about 75%. She had puzzled for some time as to whether such cognitive abilities could be said to be indicative of consciousness but felt reassured when hearing Colin Blakemore define human consciousness in purely cognitive terms. There were many at this conference, though, who would be unhappy with such an operational definition.

Paul Bloom finished the session with a paper "Language and consciousness" which was highly relevant to some of the recent debate on jcs-online. He started with Darwin's question -- is our rich mental life the result of language? -- and went on to address the theories of Whorf, Hobbes and Quine that language was essential for concept formation. Bloom acknowledged that words can motivate categorization via labelling and can give rise to distinctions that we would not be otherwise conscious of -- an example being the training of wine tasters, but felt that the strong conclusion that experience was in some way created by language was over-stated. He referred then to the observations of William James and Oliver Sacks that abstract thought was perfectly possible without words. Susan Shaller's work among deaf isolates indicated a rich pre-linguistic mental life. When they do finally learn words they pick them up very rapidly. Bloom's interpretation of the case of Helen Keller was that she learned to apply labels to pre-existing experiential categories. Noting that pre-linguistic infants are perfectly capable of abstract object individuation he concluded that language was not essential for the creation of concepts.

The next two sessions -- transpersonal psychology, and conscious vision in the striate cortex indicated the diverse range of the consciousness debate. In the first session Charley Tart presented a critique of science as a social phenomenon. He argued that science had been very successful in some areas but we needed to reconsider methodological issues before it could deal effectively with the full range of human experience. Roger Walsh then presented a brief history of transpersonal psychology, concluding that for a movement that was only 25 years' old great progress had been made.

The contrast with the afternoon session could not have been more pronounced. Three distinguished researchers, Petra Stoerig, Robert Turner and Dov Sagi all addressed the role of primary visual cortex (V1) in conscious vision. I wouldn't claim to be competent to comment on any of the wealth of functional MRI and other data presented, but was interested by the ongoing debate on the binding problem. According to Llinas' talk on Tuesday, binding was a bottom-up process, derived from temporal synchronisation among cells in diverse cortical areas. Robert Turner showed in his talk how it was inappropriate to view area V1 as some sort of passive sensory mapping of the external world as there was clear evidence that V1 is subject to top-down modulation from higher cognitive set-determining regions (typically frontal). An interesting addition to this panel was philosopher Ned Block who pointed out a confusion among psychologists with the use of the word "consciousness". Block delineated his distinction between phenomenal experience (P-consciousness) and access (A-) consciousness -- the latter dealing with the global control of action, decision making, reporting etc. According to him there was some confusion between these terms, in particular in the recent work of Crick and Koch. He suggested their claim that V1 was not conscious (because it was not connected to frontal areas) was actually dealing with A-consciousness and was at best an empty tautology. However it told us nothing about P-consciousness. Dan Dennett, who doesn't accept Block's distinction, was the first to the microphone to challenge him, but the episode showed how essential it is to include philosophers in even the most hands-on areas of brain research.

We then split up into groups for concurrent panels, with the parapsychology session attracting one of the biggest audiences

Thursday 11 April

The fourth day opened with Stuart Hameroff introducing a session on Medicine and Consciousness. Nicholas Franks spoke on Stuart's own subject of anaesthesia, giving evidence for the view that general anaesthetics act much more specifically than had previously been thought. The accompanying slides were both easily visible and very helpful in understanding the points being made. This was not universally true of the visual aids during the conference. Andrew Weil then made a plea on behalf of the body's capacity for self-healing, suggesting that clinicians-in-training spend far too much time with very sick people. This gives them a distorted picture, since these are by definition the cases where natural healing mechanisms have (unusually) failed. In consequence, the medical profession pays too little heed to the significance of their patients' states of consciousness as a way of accessing the body's natural healing processes.
The hall was full for the next session, "Quantum Processes in the Brain", which had been widely previewed as a clash between Pat Churchland and Stuart Hameroff on the theme of their JCS articles [Vol. 2 (1), pp. 10–29; Vol. 2 (2), pp. 98–111]. The session kicked off with Freidrich Beck and Dimitri Nanopolous -- each providing some fairly dense technical material which went some way over my head. Nanopoulos in particular, who was billed to talk on superstring theory -- "brains with strings attached", actually just churned out strings of illegible equations. It underlined the need for particular care when preparing to address an interdisciplinary audience.

When Pat took to the podium there was a noticeable hush in the auditorium. Starting off in a fairly subdued way, she soon started striding up and down the stage, giving a wonderful oratorical performance. It made me realize just to what extent the real superstars on the consciousness scene are there because of their personal charisma, and that presentation was just as important as content.

And what about content? Pat was billed to talk on "our friend the microtubule", but immediately launched into a 25 minute harangue of Chalmers' hard problem. And she made some good points, describing the hard problem as an argumentum ad ignorantiam with the following structure:

We don't know much about X, therefore:

X can never be explained
X can never be reduced to anything else
X can't have properties p1 . . . pn
On this view, nothing science could ever discover could deepen or revise our current ideas about X, but according to Churchland this was an elementary logical fallacy which she taught routinely to her students. It worried her that recently, whenever researchers presented a new piece of work in some aspect of cognition, at the end someone would inevitably get up and say -- OK, but what about the hard problem? This was leading to a certain explanatory reticence within the cogsci community. (Advocates of the Chalmers position might prefer to call this conceptual clarity or humility.)
At the end of her presentation she did offer a few asides on the qm theory. This time it was not caterpillars with hookahs [as in her JCS article] or pixie dust in the synapses [THES, 5 April 1996, Consciousness Supplement, p. vi] but she delighted the audience with an anecdote on the difference between experimental and practical witchcraft.

It would take a brave man to follow up on Pat Churchland, and that could only mean Stuart Hameroff. His presentation, in contrast to the preceding speaker, gave more weight to the scientific issues than the presentation. Having disposed of some of the criticisms of the microtubules theory which had come from Churchland and Dennett, he went on to give a good account of Penrose's theory of orchestrated objective reduction [published in JCS Vol. 3 (1), pp. 36–53].

Friday 12 April
This morning Dave Chalmers displayed his prototype "consciousness meter" for the first time during a debate with Paul Churchland. (OK -- the thing looked like a hair-dryer but it can't have been, as the qualia-detection light, functioning when the machine was directed towards ordinary mortals, only flickered when pointed at Pat and was resolutely off when aimed at Dan Dennett . . .) The humorous aspect aside, Chalmers' presentation was a serious effort to show that he is not simply "Mr Hard Problem" but has other concerns as well, with a plea to clarify the implicit philosophical assumptions underlying the search for the neural correlates of consciousness.

Paul Churchland's presentation was one of a number during the conference which illustrated a problem with well-known and busy speakers: the danger of "the standard talk". It was very interesting to those hearing it for the first time, but for too many in the audience it was the second, third or nth time -- right down to the slides. It has been suggested that for "Tucson III" we should leave at least half the plenaries open for unknown people with interesting ideas and only accept "name" speakers if they undertake to put more original material into their presentations. It has to be said that Tucson II was more a place for consolidation than for groundbreaking new theories. This was quite appropriate on this occasion, in order to get the field established, but no one should think they can rest on their laurels for the next conference.

Most of the experimental work on consciousness is being done on vision, for reasons laid out clearly by Francis Crick in his book. Having addressed area V1 on Wednesday the spotlight of attention (sic) now switched to the extrastriate cortex, with masterful presentations from Christof Koch, Roger Tootell, Nicos Logothetis and Larry Weiskrantz, presenting a wealth of evidence and theories on vision and its pathologies.

For me, the highlight of the conference was the panel on Phenomenology and Experiential Approaches. This panel had been very carefully structured and began with John Searle and Max Velmans exchanging views on scientific method. Although they disagreed on certain details (is a pain in the finger actually a pain in the head?) they both agreed that the hard line between subjectivity and objectivity was not justified. Searle pointed out that although consciousness was ontologically subjective this didn't mean that it could not be accessed and studied in ways that were epistemologically objective, given sufficient methodological rigour. Francisco Varela filled in some of the detail on this with a paper on "Neurophenomenology", pointing out that the tradition of Husserl provided a toolbox for the study of subjective states, but that this required many years of rigorous training.

I got the feeling that his interest in phenomenology was partly fuelled by many years of meditation practice but this is reading between the lines, and was not explicitly stated. However, our own Bob Forman was quite explicit on this matter, arguing that scientists like to study nature in its simplest form and then extrapolate to more complex systems. So, in the same way that a biologist might chose to study the paramecium or E coli, the study of consciousness should be based on what he called the pure consciousness event (PCE), which many mystical traditions argue is the ground state of conscious awareness.

Although Forman and Searle are approaching consciousness from radically different perspectives (with Varela and Velmans somewhere in between) there was remarkable agreement on the scope and methodology that was appropriate for the investigation of consciousness. Searle commented that since the 17th century we have lived in a demystified world and as such, in principle, nothing is beyond the reach of scientific enquiry -- what is important is the intersubjective rigour of the methods. As consciousness is puzzling, we should not rule out any approach to gaining understanding a priori.

Saturday 13 April (by Jean Burns)
The morning session was on parapsychology. This topic was not presented at Tucson I (except for one poster ), and even at Tucson II was deemed sufficiently controversial to require a representative "sceptic" on the panel of four, to ensure a balanced presentation. The sceptic was Susan Blackmore, a former parapsychologist, and the other panelists are all actively involved in experimental parapsychology.

Psychologist Daryl Bem started, with a description of the extensive studies of ganzfeld experiments, in which a subject hears white noise and has ping pong balls (halves) taped over his eyes. After relaxing, the subject reports his mental impressions for 30 minutes, while a "sender" gazes at a film clip, or something of the sort. Without going into the details, a judge who does not know the target compares the report narrative both to the film clip used as target and three other clips (all chosen beforehand), and the best fit is chosen. Analysis of these experiments show a statistically significant correlation between the target and the "best fit". The experimental procedures and methods of analysis have been examined and debated for a number of years. Bem (who is also a magician who routinely fakes mind-reading as an entertainment), was originally sceptical, but became convinced of their validity.

Susan Blackmore's story was somewhat different, since she started out believing that parapsychology could help her understand the nature of mind and memory. But after a time of unrewarding experiments of her own, together with suspicions about the results of another experimenter, she went on to generalize that psi effects are unlikely to exist. She also claimed that in any case psi has nothing to do with consciousness. I have always thought this assertion was preposterous, but I like such claims if the propounder can make a good argument, and found Sue's arguments delightful -- basically saying that if these effects exist, they might be explained by non-conscious processes.

Roger Nelson presented results of recent experiments which showed that at times when many people in the world have their attention on the same thing, as at the time of the OJ Simpson verdict, random noise generators develop correlations and are no longer quite random. Besides the Simpson verdict, he showed this effect occurs during any peak time television programming, but not during the advertising periods.

Dick Bierman showed that if the experimenter has the subject hold an intention in a PK (psycho-kinesis) experiment, even if the target is randomly chosen after the experiment has been done, significant results are nevertheless obtained. Dick suggested that the explanation could be related to the well-known "experimenter effect" in certain interpretations of quantum theory.

These presentations were a little technical, with a great deal of discussion of methodology, and I'm not sure the audience really absorbed the descriptions of the above experiments. Marilyn Schlitz, who was moderator of the group and a member of the planning committee of the conference, stopped by while I was writing this and made much the same comment. But the experiments are fascinating and I am glad they were presented.

In the late morning session, V.S. Ramachandran gave a presentation on patients who suffer a severe stroke in the right hemisphere and experience "neglect" of the left side. Even though a patient's left arm is paralyzed, the patient will vehemently deny that the arm is paralyzed at all. Ramachandran explored this with a number of patients. For instance, he asked a patient "Will you clap your hands?" The patient clapped with one hand, while claiming she was doing it with both. (Ramachandran said we now know the sound of one hand clapping!) In a bizarre move, Ramachandran poured ice water into the ear of one of the patients. It seems that ice water in the ear causes rapid eye movement, and, astonishingly, after the rapid eye movement, the patient then knew she was paralyzed. The effect of "neglect" can occur with stroke of the right hemisphere (with paralysis on the left side), but never with stroke of the left hemisphere (with paralysis on the right), which led Ramachandran to a fascinating link with Freudian denial and repression theory .

D.L. Schacter talked about memory. He read a list of words to the audience, which were along the lines of sugar, sour, taste, honey, and the like. He then asked how many remembered the word "taste", emphasizing that he was asking about a specific recollection that the word was on the list. About half the audience raised their hands. He then asked how many remembered the word "sweet" in the list. Almost as many raised their hands. But "sweet" wasn't on the list! He then went on to explain this in the context of his theory of implicit memory.

Finally, we had the physics talks. Henry Stapp said consciousness can be explained by quantum mechanics, the reason being that the formulation of quantum mechanics inherently refers to conscious experience. Avi Elitzur suggested that current physics does not seem able to adequately discuss consciousness. He pointed out that in the block universe of relativity, all events which happen to a person must be considered to exist on a worldline, with conscious experience simply moving along the pre-existing worldline. Furthermore, conscious experience of different people does not necessarily move in synchrony. So when you talk to someone else, that person might simply be a zombie, with their conscious experience having moved on. But he finds this view unbelievable, and concludes that instead a paradigm change is in the offing in which our understanding of the laws of physics will be expanded.

Al Scott, from the conference program committee, ended the conference with a plea for the study of consciousness as an emergent phenomenon spanning every level from quantum physics to culture and society.



A final word
Keith Sutherland
Perhaps the most remarkable -- and creative -- thing about the conference was (like the jcs-online list) the presence of people with starkly differing intuitions and backgrounds debating the subject together. I was in the middle of a conversation with Bob Forman (Religion, CUNY), when Robert Turner (Institute of Neurology), who had earlier presented his MRI research at the V1 symposium, interrupted us to draw Bob's attention to the writings of the Sufi divine Ibn Arabi. Christof Koch (who normally gets to hang out with macaque monkeys) spent much of his time at Tucson talking to the transpersonalists. This summed up the spirit of open but rigorous enquiry that has been championed both by the conference and the Journal of Consciousness Studies.