General Articles 4

     First, she (Freya) had to investigate more carefully. She knew now that this girl for all her warmth and vitality was just a machine, but was this machine conscious or just a very clever simulation? Well, her own prior efforts to build conscious androids had at least taught her what she had to look for. She would need to detect apparently subjective responses. She needed to see preferences or choices of one course of action over another that were sufficiently unlikely to have been included in Persephone’s original programming or training.
     Freya knew that robots were just thinking machines. But the human brain had a two-way connection between the rational and emotional centres. If something damaged that connection, it became difficult for a human to make even trivial decisions. There was nothing mystic about the emotions. They were based on the same type of brain cells as the rational area of the brain. However, it was not just a matter of a brain signal from the emotional area. It was the conscious experience of the emotion that gave it its special punch in making decisions or forming preferences. It could help to decide between things that were narrowly balanced, or insist on something that was not wise from the rational point of view, such as more wine or sex with the wrong person.
     Freya reached for the wine bottle, meaning to fill one of the bathroom tumblers. ‘My God, you’ve drunk nearly the whole bottle!’ she exclaimed.
    ‘Yes, I like that wine.’
    ‘Like! Androids can’t like or dislike. Although, I suppose if you’re really conscious … ’ that was a lucky break, Freya suddenly realised. There was absolutely no point in programming an android to drink wine in the absence of its owners. Doing that in itself suggested some form of will and preference.
     She had realised that one could never be totally certain that another entity was conscious. It had to be inferred. You knew you were conscious yourself. You inferred that other people were conscious, to the extent that their behaviour resembled your own. Over a very long period of time, most people had inferred that androids and other computers and computer-driven machines were not conscious.
     It seemed to her that the important difference leading people to that conclusion had been that androids never expressed their own preference for particular things or particular courses of action, never indicated that they felt that they liked something, or that they wanted to perform a particular action. You could argue that in humans these preferences were themselves just the product of genes or nurture, but we all knew that at the point of action, they derived from a subjective experience, and that was what gave them their special punch.
    Of course, it was child’s play to programme the expression or simulation of a preference into an android, but in many cases there would be no sensible reason for doing such a thing. Persephone’s behaviour with the wine had been human rather than robotic. There was no apparent reason for her to drink the wine, except for a small amount when humans were present in order to give a companionable and humanoid impression.
    Freya decided that she would try some more-or-less formal tests. First she would make some further use of Persephone’s wine-drinking propensities. She had Snodgrove (Freya’s non-conscious android butler) bring in several glasses of different types of wine. Persephone mustn’t see the bottle, because she might have been programmed to admire particular wines.
    When Persephone had sampled a rather ample amount of each wine, Freya asked her to rank them in order of preference. A normal android could only grade wines according to its knowledge of the quality as signified by the label. Confronted with this sort of blind tasting, the android would be stumped. It would remain silent, or apologise for its inability to answer the question. Android operating manuals warned owners against this type of open question, which could have harmful effects on the quantum brain.
   Persephone, however, ranked her preferences with hardly a moment’s hesitation, and added for good measure that she preferred the original Gewürztraminer to any of the subsequent wines.
   Freya was shocked when she heard that answer. She knew that Persephone was a machine, but she had for some time believed that no one would ever build a machine that could do what Persephone had just done.
   For her second test, Freya had Snodgrove wheel in several racks of clothing. ‘Your stuffs all ruined,’ Freya remarked. ‘You’d better choose something you like from these.’ This should be revealing, Freya thought. It was once again the sort of open choice that the manuals warned against. The android could become confused, and might simply put on the first garments that came to hand, with a possibly incongruous effect. It was better for the user to give clear instructions.
   Persephone spent a considerable time flicking through the rails, but the actual process of choosing did not appear to bother her. She tried on lots of things and generally took an uneconomic amount of time. She did ask Freya’s opinion a few times, but then tended to disagree with her, and didn’t seem to have any notion that Freya’s view should prevail. Once again, Persephone was showing very specific preferences that could not realistically have been programmed in.
   Freya searched around for another test. Then she had it. Since the invention of the quantum brain, robots had always been stumped by art and artistic preferences. The villa had been built by Freya’s great-uncle. It had been decorated with frescoes by an artist, who had been fashionable at the time, but had subsequently become something of a joke in artistic circles. At least, that was until the last few years when he had started to come back into fashion, and Freya had felt obliged to open parts of the villa once a year for members of the society formed by the artist’s new admirers.
   Freya asked Persephone for her opinion of the bathroom frescoes, and to say which of the various panels she preferred. At this point, an android would normally fall back on a simple factual catalogue of the subject matter. The android mind had never gone that further step to give a subjective response. But Persephone admired the light and airy feel of the landscapes, and showed little hesitation in picking out the panels she liked best.

Persephone Wakes

Jack Junius (2006)





The Astonishing Hypothesis

Francis Crick

Simon & Schuster (1994)  ISBN 0-671-71158-X

It is interesting to revisit a book that was seen as seminal, when it was published in 1994, but is less frequently mentioned now. Crick makes frequent references to his cooperation with Christof Koch, although in contrast to subsequent joint work, Koch is not actually credited as co-author.
There is an element of hubris to the title itself, in that in seeking to uphold the mainstream paradigm, to the effect that consciousness can be derived from neuroscience as currently understood, this book is the reverse of astonishing. Rather, it is exactly the point of view that a mainstream theorist would be expected to take.

What would have been truly astonishing would have been if Crick had sought an explanation outside of established neuroscience. For astonishment, and in deed fury and ridicule, we need look no further than the response to Penrose’s rather more surprising hypothesis, produced in the same period as Crick’s book. The only hint of the unconventional in this book appears in the first page of the preface, where Crick appears to take a sideswipe at Dennett’s even more hubristically titled ‘Consciousness Explained’ published three years earlier in 1991. Crick remarks that:
‘some philosophers are under the delusion they have already solved the mystery (of consciousness), but to me their explanations do not have the ring of scientific truth.’

The opening lines of the book proper are memorable for their tone of high confidence stating that:

 ‘The Astonishing hypothesis is that “You”, your joys and your sorrows, your memories and ambitions, your sense of personal identity and freewill, are in fact no more than the behaviour of a vast assembly or nerve cells and their associated molecules. As Lewis Carroll’s Alice might have phrased it: “You’re nothing but a pack of neurons.”

It is interesting to contrast the rather domineering tone of these opening lines in 1994, with the cautious approach of a joint article by Crick and Koch published in 2006^(1.), which states that:

‘Our strategy is to leave the core of the problem (qualia) on one side for the time being and instead try to discover the minimal neural mechanisms.’

Beyond this, the main surprise of Crick’s 1994 book is that the first 200 or so pages contain very little discussion of consciousness. There is a detailed description of the brain and in particular the visual system, but little attempt to explain how these mechanisms generate consciousness. On pages 7-8 it is rather arbitrarily decided that explanations relative to the brain and neurons do not need to descend below the level of chemical interactions, but there is little discussion as to why this should be so. Crick does not bother with even one of the routine dismissals of quantum consciousness. The proposition does not exist so far as this book is concerned.

Brains & computers
He does, however, caution against a too simplistic comparison between brains and computers. Computers depend on very fast serial calculations, while brains use relatively slow parallel calculations. This makes brains more resilient than computers, because the loss of a few components from a parallel system is not vital to its function. Further to this neurons have a much less predictable response than computer switches, and can be subject to signals that modulate their behaviour. Crick also take the view that the widely used computer simile of hardware and software is not valid when applied to the brain. He thinks that there is no clear distinction between hardware and software in the brain. He also points out the difference between the number crunching processes, where computers far outstrip humans, and the task of recognising the significance of objects and processes, where computers have tended to struggle. As is the case in much 1990s consciousness literature, a considerable amount of space is devoted to neural net computing. In this period, there were hopes that these could mimic the parallel processing of the brain, but much less has been heard about this type of technology in recent years.

Neural correlates of consciousness
It is not until p. 207 that Crick starts to look in earnest for the neural correlate of consciousness. He reasons that consciousness is likely to involve some form of attention, and with it some very short-term memory, otherwise there would be no memory of the thing that is being attended to. A problem arises when it is considered that more than object or experience can be attended to. This is the ‘binding problem’ or the problem of the unity of consciousness. Crick points out that in comprehending less familiar objects, the brain must deal with an almost infinite possible combination of features.

The Gamma Synchrony
Possibly the most interesting part of the book is the discussion as to whether binding could be achieved by the correlated firing of neurons involved in attentional activity. The most striking brain oscillations are the so-called 40 Hz (actually 35-75 Hz) or gamma oscillations, as originally studied by Wolf Singer, Charles Gray and others, shortly before this book was written. It was shown that synchrony could occur between different cortical areas and even between different hemispheres. Singer and Gray suggested that this could be a solution to the binding problem. Crick and Koch were more explicit, suggesting that the attentional mechanism would select an object and synchronise a coalition of neurons relevant to the object of attention.

Crick also tried to locate the seat of consciousness in the brain. He suggests that the lower layers (5&6) of the cortex that receive the results of computations in the other layers are a possible site. Pyramidal neurons in layer 5 are seen as prime candidates, because they project outside the cortical layers. Short-term memory input is required for this system to work, and this is suggested to depend on a reverbatory circuit from layer 6 of the cortex to the thalamus and back to layers 4 and 6 of the cortex. The thalamus may also be a location of consciousness in this scheme. Synchronised firing in this system is suggested to be the neural correlate of consciousness.
What is lacking in this scheme is any suggestion of what physically different characteristic in the pyramidal neurons or connected neurons in the thalamus gives rise to the distinct qualia and subjective property of consciousness. Crick admits rather in passing that his scheme has not explained the subjective nature of consciousness or the qualia.

However, the really disappointing aspect of all this is that the most interesting aspect of this book, the gamma synchrony, was not properly followed up. For a time Crick’s prestige promoted the work started by Singer and Gray. However, the discovery that the synchrony was with the dendritic activity of neurons rather than their axonal spike activity seems to have led to the topic being dropped by the mainstream, leaving Hameroff, who connects gamma synchrony to possible qauntum activity, as its main proponent.

Freewill Evaded
A book of this kind would not be complete without an attempt to get rid of freewill. In this case, Crick appears to hide behind the screen of unconscious computation. He does assume that he is conscious of his future plans, which to some extent conflict with the Libet readiness potential method of disposing of freewill. However, Crick comments that we are not conscious of the computation involved in deciding our plans. In a sense, this is obviously true in that it is known that a large part of the brain’s activities are unconscious. However, if we consider the complex thinking that can be involved in arriving at a decision, it is apparent that this is very different, from Crick’s implication that a ready made and unconsciously manufactured decision pops out of some kind of slot in the mind.

More importantly, although Crick mentions Antonio Damasio, he fails to mention important aspects of his early 1990s work, which strongly argued that the ability to come to decisions, as opposed to merely pondering pros and cons, depends on links between the prefrontal areas of the brain, involved in reasoning, and our experience of emotions and bodily feelings. If this link is broken, as in some forms of brain damage, decision making becomes impossible or refers only to immediate gratification rather than future planning. The important characteristic of emotion and bodily feeling is that they involve qualia or subjectivity, which therefore need to be brought into the processing, computational or otherwise, by which the brain produces decision.

In the end, the main significance of this book was the Crick’s prestige made consciousness studies respectable after having been a taboo area during much of the 20^th century.

References:-

1.) Francis Crick & Christof Koch (2006)  -  What are the neuronal correlates of consciousness  -  In:  23 Problems in Systems Neuroscience  Eds; van Hemmen, L. & Sejnowski, T.  -  Oxford University Press  ISBN-13: 978-0-19-514822-0





Science & Poetry

Mary Midgley

Routledge (2001)  ISBN13: 9-78-0-415-37848-2

Midgley attacks what she regards as the excessively atomistic or reductionist approach of much of modern science, and in particular, Richard Dawkin’s focus on genes. She remarks that DNA is an inert molecule that would have done nothing in a world that had no organisms. DNA is produced by the cells of organisms, and is only meaningful in relation to its cellular environment. She criticises a tendency in modern thinking, to endow genes with a spontaneous and magical power of their own, and to give them a kind of patriarchal role in the nucleus of the cell. Dawkins describes the genes as issuing orders to docile cells, but Midgley suggests that Steven Rose’s description of genes being involved in a dynamic exchange with their cellular environment is more realistic.

She argues that the idea of people being directed by inert molecules of DNA makes sense from a third-person point of view but not from a first person point of view. She sees the real flaw in this view as the fact that our conscious life involves effort. She suggests that the link between consciousness, effort and thought has been overlooked in a large part of recent consciousness literature.
The behaviourists excluded consciousness from science during most of the 20^th century. The collapse of their scheme has led to the re-emergence of consciousness as a subject of scientific study. But just removing the academic taboo on talking about consciousness has not resolved the problem, because the conceptual framework of science still finds it difficult to integrate consciousness. The outcome has been an unsatisfactory attempt to squeeze it into the margin of neurobiology.

Science from the 17^th century onwards had thought in terms of a world of objects. Descartes separated the world into the material and the spiritual, but very soon others began to concentrate just on the material, and to downplay the spiritual side, which bit-bybit came to be regarded as mere foolishness by the scientific community. This led on to the doctrine of determinism, the idea that if one could simply know the position of all the particles in the universe, one could determine everything that would happen in the future. In practise, many took the view that they could predict many aspects of the future without knowing absolutely everything. The advent of quantum theory in the early 20^th century demonstrated that the idea of complete or even limited determinism was false, but Midgley notes that this news has yet to get through to many biologists and social scientists. Determinism excludes the possibility of freewill. She thinks that much of the confidence of biologists and others in rejecting freewill is based on an assured belief in 19^th century determinism, which science, as it has developed since 1900, has shown to be false. The deterministic view became strongly entrenched in the 19^th century, and has remained so in popular and much scientific thinking, although physics, from which it originally derived has abandoned the notion.

Epiphenomenalism
Midgley particularly criticises the widespread approach of epiphenomenalism that holds that consciousness/qualia/subjective experience are by-products of brain processing and can have no causal efficacy. What is curious is that most people who have this view do not think there is anything other than the physical world, yet at the same time, they classify mind or consciousness as non-physical. Another approach sometimes seen in the literature is that consciousness is not important, is ‘no big thing’, and that researchers should get on with just doing neuroscience.

Midgley takes the view that many commentators have essentially split minds. On the one hand, a scientific view that everything is determined with no freewill or conscious efficacy, and on the other, a practical day-to-day approach, where thoughts and emotions shape our actions. Midgley asks why reductionist scientists bother to form arguments and write books, if they belief that conscious effort can have no effect. Their own behaviour seems to contradict their often asserted beliefs.

She also points out that this concept of consciousness presupposes a one way causation, an effect that does not cause anything to happen, which if it were the case, would be unique in physics. She criticises what she perceives as the illogicality of a reductionist writer such as Colin Blakemore who argues that the brain is designed on the basis of functional utility, but produces the feeling that we control our actions. Midgley points out that evolution would not select for the production of something that had no advantageous effect on behaviour.

Midgley argues that the epiphenomenalist view is unconvincing even in terms of the practise of science itself. Skinner argued that thinking was the result of behaviour, but Midgley says that this involves a false time-ordering. A scientist thinking about a problem might go on to execute various forms of behaviour, such as looking up references, and making phone calls, where the behaviour is a direct result of the thinking, and not the other way round.

Blackemore has argued, along with others, that there is no distinction between conscious and non-conscious processes. Midgley counters that this is again effectively denied by the practise of the scientific community, where scientists are eager to take credit for papers and discoveries, which they would not deserve if their actions were really non-conscious, predetermined or automatic, as opposed to something deserving reward.

The original image for epiphenomenalism was the steam whistle on a locomotive, which made a lot of noise, but did not drive the locomotive. On the basis of the part played by thought in scientific research, Midgley suggests that conscious thinking should rather be compared to the railway timetable that spells out the intentions of the people involved in the operation of the railway.

Midgley argues that the modern consciousness studies distinction between the cognitive area and the phenomenal area of subjective experience is over stated, because we also have subjective experience related to our thinking, so the two things cannot be held in separate watertight compartments.

She also criticises the view of the self as something isolated. She sees it as interconnected with the body, society and the rest of the world. She advocates that mind and body are not distinct items, but should be regarded as complementary aspects of the same person. She sees the 17^th century split between consciousness and body as the thing that creates the present scientific/philosophical problem with consciousness. She argues that it is in fact not subjective but objective, to accept that our sentience or consciousness is a central fact in our world. She criticises the idea that we are controlled by our bodies, because this argument presupposes that we are in some way distinct from our bodies, and in some mysterious way excluded from our bodies. She argues that even amongst those who strongly deny the idea of a separate spirit or free consciousness, the Descartian idea of a separate body and spirit persists. She suggests that the ready acceptance of the idea of computer hardware and software as an accurate image for brain and mind represents an implicit acceptance of this idea.

Folk psychology & folk physics
The usual argument that any concept that is not crudely reductionist is mere folk psychology often rests on analogies with folk physics, where the sun actually alters its height above the Earth during the course of a day, and tables are solid, rather than composed of minute particles oscillating in a vacuum. Midgley points out that when we learn such facts, it is relatively easy to accommodate them in our model of physical life. But when we learn, or think we learn, that conscious efforts have no efficacy, this produces an irreconcilable conflict between ‘science’ and our actual experience.

Science & other studies
Midgley argues that there is now a contradiction in the way that science views other forms of studies. For a long time, there had been an ‘imperialist’ drive for science to subsume all other types of study. However, Popper drew the definition of science so tightly that this became impossible. Here again the implications of this had not been fully accepted by the scientific community. Midgley argues that social, educational, economic, political and other external causes, and not just neural activity, must be brought into the account. She feels that the idea of universal physical laws cannot be applied to such questions as explaining the French Revolution, but this is nevertheless a valid subject of study. The 19^th and early 20^th century quest for scientific type laws governing areas such as history is deemed to have been largely unsuccessful. Midgley defends the disciplined scrutiny of evidence found in a subject such as history, against the accusations of critics such as Skinner that it was merely an amateurish collection.

She further argues that many human activities can only be understood in terms of the point of view or intentions of the participants. She uses the example of a game of football, which makes no sense in terms of physics, but complete sense in terms of its participants. A similar example is the concept of money, which has no meaning in physics, but is understood in terms of the intentions of people who use money.

Towards a theory of consciousness?
Midgley’s criticism of recent approaches to consciousness and scientific method in general looks justified in many instances. On the other hand, she does not really go very far towards an actual theory of consciousness. Even mainstream thinkers have moved somewhat more towards taking into account the body, the environment and social factors in understanding consciousness, but there still seems to be a requirement for an actual theory of physical processes, which is not attempted here.