- Introduction: we want to analyse the phenomenon of consciousness into something more tractable. Perhaps tabulation is the way forward
- Some grounding: a reminder of what it is we want to tabulate. What a frame of consciousness might look like
- Some useful measures of consciousness: some one dimensional measures of consciousness which are useful in hospitals
- Descriptive experience sampling: a technique for sampling the contents of consciousness
- Motivation: the four dimensions of the authors of the prompting papers, that is to say references 1, 2 and 3. One dimension is not good enough
- Hierarchic classifications. A popular but presently inappropriate way to analyse complex populations
- Cross tabulations. A recap on what cross tabulations do
- A recap on the reference 4 collection of variables
- Why are cross tabulations useful here? What sort of things are cross tabulations used for? Does that read across to the present case?
- Some complications: some variables are more tricky than at first might appear
- Four dimensional graded consciousness: the four dimensions variously proposed in references 1, 2 and 3. The argument for dimensions is made; the argument for these particular dimensions is not so made
- Conclusions: some cross tabulations of normal frames of consciousness would be good. Which might, inter alia, tell us which variables, which dimensions are interesting or important
- References
Introduction
When I was young, I spent quality time on producing cross tabulations about various large groups of people, perhaps all the employees in employment in England and Wales. This at a time when there were government statistical organisations with a brief to produce such things and when being an employee in employment, with a National Insurance card, was the normal state of affairs for normal people. Plus regulation hand towel, tablet of soap and ashtray in the case of civil servants. This is reflected in some of my blog posts over recent years.
But this post is prompted by the papers at references 1, 2 and 3 which, amongst other things and in amongst a fair amount of philosophical flavouring, promote the idea that consciousness can usefully be analysed into four dimensions, papers which we eventually reach in what follows in the section on four dimensional graded consciousness. Two of our efforts at such analysis are to be found at references 4 and 5. Reference 6 is rather different, but also relevant.
In which connection, we note that quite a lot of people believe that that human feelings, that emotions, can be usefully analysed onto a two dimensional half space, with one dimension being arousal, say a real number in the range [0, 1] and the other being valence, say a real number in the range [-1, 1], with minus one being bad and plus one being good. With there being a very real difference between the first half range and this second, full range, where the zero is a pivot rather than a beginning – although some blur this difference by measuring arousal above and below some mean level. Not to mention all the statistical people that like to extract a couple of principle components – that is to say dimensions - from some great mass of data. Plotting things on one side of a piece of paper has a lot to be said for it.
One of the other things in our three papers is the assertion that consciousness is indeed a singular phenomenon, eventually to be accounted for by a singular bit of neural machinery. Another is that it is a graded phenomenon, that it is not a matter of on or off. One consequence of which is that there are likely to be marginal cases out there, both among us humans – thinking here of differences between humans and of changes in any one human over time – and in the animal kingdom at large – thinking here of some animals being conscious in some way and other not. We agree with both these assertions and we do not see a problem with the consequence.
In any event, patrolling that particular boundary, those marginal cases, is not our present concern. Our present concern is cross tabulations of frames of consciousness (of LWS-N, on which there is introductory material at references 19 and 20) and the four candidate dimensions - and our suggestion is that cross tabulating populations of normal frames of consciousness would be a useful adjunct to looking at abnormal frames of consciousness.
For the avoidance of doubt, we try to reserve the word ‘subject’ for the person (or animal) who is conscious, who experiences the frame of consciousness, and the word ‘object’ for what that frame of consciousness is about.
Some grounding
We start by grounding the discussion which follows with a description of a take containing several, perhaps many, frames of consciousness. A description which is intended to illustrate the potential complexity of the contents of consciousness – while not forgetting that in many cases, that for much of the time, the contents will be much simpler, minimal even.
This take is the three minutes it takes to brush my teeth morning and evening, for which purpose I use a brush, rather than a machine, and an egg timer. Note that this has been a twice a day activity for years: same place, more or less the same times, every time: the brain is able to get on with it, leaving plenty of room for other stuff to be going on at the same time. We do plenty of stuff for which this is not true. Looking forward to Hurlburt, an oral primary activity with mainly oral secondary activity.
Standing in my dressing gown at the sink in the bathroom, left hand on the window sill behind the sink to maintain balance and position. Eyes mostly shut, but opening from time to time to check progress of time on the egg timer. In any event, the window glass is frosted so there is not that much of interest to be seen.
Pressure on the left hand on the window sill.
More complex tactile sensations from the right arm and the right hand holding the brush.
Pressure on feet. Perhaps some tactile sensations from foot movements, perhaps stretching up on my toes from time to time to make a change from carrying my weight on my heels.
Dressing gown flapping a bit against legs.
Sounds from the brush against teeth and gums.
Sounds from the cleaning stuff sloshing around the mouth.
Occasional sounds from the blobs of cleaning stuff dropping from the mouth into the basin of the sink.
Tactile sensations from the lips, mouth, tongue, gums and cheeks. Brush moving around, from place to place in the mouth. Some reciprocal tactile sensations from the active tongue brushing against soft parts of the mouth.
Taste sensations from the tongue.
Burning sensations from the cleaning stuff on the inner lips – burning is not quite the right word here, but a better has not come to mind.
Occasional abdominal gurgles of one sort or another.
Pleasant surprise when time on the egg timer seems to have leaped forward.
Mild irritation when time on the egg timer seems to have stopped.
Inner thoughts about the upcoming visit to Hampton Court Palace. In words but not articulated out loud – which would, in any case, be difficult in this context.
Inner thoughts about the gurgling radiator behind me which needs to be bled.
Hard to pin down exactly which elements are in any one frame of consciousness, but my belief is that it is more than one thing at a time, more than one strand at a time, even though most of one’s attention is on just one thing at any one time. Something which exploits the layers of LWS-N. And something which it is easy enough for the interested reader to try for himself; to explore the potential riches of the conscious experience!
Some useful measures of consciousness
Doctors need tools to help them assess the state of consciousness of people who, for one reason or another, are not very responsive. Ideally tools which deliver simple numeric scores, with low scores for the unconscious and high scores for the conscious, a simple one dimensional view of consciousness which can help making clinical decisions. Simple numeric scores more like those for a variable like body temperature than the categories of something like the demographic variable marital status. Three examples of such tools follow.
The first is largely a matter of observation, the second involves EEG electrodes and the third requires rather more than that.
Glasgow coma scale
 |
| Figure 1 |
The answers to the three sections – eye opening, verbal response and motor response – can summarised in the form of a numeric score in the range 3 to 15, with low numbers bad, high numbers good.
Bispectral index
 |
| Figure 2 |
The BIS Index is a number between 0 and 100, scaled to correlate with important clinical end points and EEG states during administration of anaesthetic agents.
Perturbational complexity index
From reference 12, we have it that the perturbational complexity index (PCI) is calculated by perturbing the cortex with transcranial magnetic stimulation (TMS) to engage distributed interactions in the brain (integration) and then compressing the spatiotemporal pattern of these electrocortical responses to measure their algorithmic complexity (information) … PCI reliably discriminated the level of consciousness in single individuals during wakefulness, sleep, and anaesthesia.
The PCI is a number between 0 and 1, with the unconscious around 0.2 and the healthy conscious around 0.6.
PCI is said to reflect the joint presence of integration and differentiation, a formulation familiar from the integrated information theory with Φ (IIT) from Tononi, one of the authors of reference 12.
Descriptive experience sampling
Back in 2017, we did some work on Hurlburt’s descriptive experience sampling (DES) with some report being given at reference 15: an interesting attempt to analyse the experience of consciousness from the inside, rather than from the outside, as in the previous section. Our work produced a sample of 1,001 frames of consciousness; a not very good sample but one which did serve to demonstrate the sort of thing that there was and from which the two tables which follow are taken.
Table 1 tells us what was going on at the sample points. With the highlight being that one was doing something purposeful nearly all of the time – but for half of that time one was doing something else at the same time, most often some kind of inner thought. We distinguish the primary activity and the secondary activity. The teeth brushing take described above is such an activity, although in that case the secondary activity was more sensational than verbal.
 |
| Figure 3 |
 |
| Figure 4 |
Inner words is when one is thinking in words, quite possibly articulating the words as far as the mouth but certainly not sounding them – except in moments of weakness – or in what is commonly called the first sign of madness. Again, inner words involves the passage of time. Thinking in pictures is less common, certainly in this sample.
The distinction between sensations and feelings is not a clear one, but the idea is that sensations is things like feeling the wind on one’s face, while feelings is things like being cross that one has missed the bus.
Neither of these tables depends on straightforward, common sense variables of the sort used, for example, in the cross tabulation of area by tenure of Figure 10 of reference 16. Nevertheless, we believe that any comprehensive analysis of consciousness ought to bring out their features, one way or another.
We also believe that the numbers in these tables are going to vary in a significant way from person to person and from time to time.
Not work which we, in the event, took forward, but perhaps there is hope yet at reference 18.
Motivation
The authors of the papers at references 1, 2 and 3 are not satisfied with one dimensional, linear, descriptions of consciousness, such as those of the section before last. They want more and they suggest that a multi-dimensional approach might better support research into consciousness.
They have come up with four dimensions, each reflecting some important aspect of consciousness, thus mapping consciousness into a four dimensional space, where each dimension might be linear, but in combination are clearly non-linear. They move on from having just one variable, one number to describe the state of consciousness to having four.
They are also interested in the question of whether consciousness is graded or discrete; put crudely whether it is indexed by a real number in the range [0,1] or whether it is binary, that is to say zero or one. We are with them in the graded camp.
The word graded carries the suggestion that each of the four dimensions was ordered, was linear, with some kind of natural progression as one moved from one end of the dimension to the other, with some sense of the distance between one point and another. Our view is that this is going to be true of some variables describing a frame of consciousness but not of others.
From all of which we take the idea of mapping any particular frame of consciousness, the subjective experience of a single person over a short period of time, typically a second or so, into a multi-dimensional space – and then aggregating the results for lots of frames of consciousness into cross tabulations.
Note that we do not yet have the option of getting the computer to identify the dimensions for us, after the fashion of, for example, principal component analysis. We do not have, as far as we are aware, the sort of data needed to feed the computer – although, as suggested in the previous section, the data one gets from Hurlburt’s descriptive experience sampling might serve. A notion that we have so far failed to properly follow up.
But before we get onto references 1, 2 and 3 proper, we look at hierarchic classifications and then cross tabulations, this last building on reference 16.
Hierarchic classifications
Hierarchic classifications are a well established way of analysing some complex whole, for example disease, industry or occupation, into more tractable parts. The plant and animal kingdoms have been classified hierarchically for a long time.
Hierarchic classifications often involve a hierarchic numeric code, for example those associated with the Dewey Decimal Code, used by generations of librarians to classify books. Sometimes such numeric codes involve or imply an order of sorts and it may be roughly true, for example, that low occupational codes are high status and high occupational codes are low status. More usually though, the order implied by these numeric codes is largely arbitrary, and the fact that one thing comes after another tells one little – apart from where to look on the shelf.
 |
| Figure 5 |
An example which yields at least two linear descriptions: one by going around the circle and the other by mapping lineages onto dates of first appearance – with the former being one-to-one and the latter being very much many-to-one. We have not tracked down the detail, but it seems likely that clockwise from east is very roughly in order of first appearance.
The evolutionary approach provides a strong rationale, a strong motivation for the construction of trees, with some purists insisting on binary divides at each stage, one lineage dividing into two.
One could in principle, if the state of knowledge were a lot more advanced than it is, do something of the sort for consciousness, mapping out the evolution of consciousness and its branching out into all different kinds of consciousness – perhaps a hundred or so, allowing for different kinds of consciousness in different groups of animals, rather than hundreds of thousands. Although in the case of higher animals the analogy is weakened by the facts that while an animal can only belong to one species, leaving aside taxonomic oddities and errors, humans at least seem to come with several different kinds of consciousness. In any event, while such a map might be both interesting and useful, for the time being it is not available to us.
And the evolutionary approach makes sense in other domains, such as occupation, where the progressive specialisation of occupations provides a natural hierarchy.
 |
| Figure 6 |
In what follows, we take the view that a dimensional approach to the analysis of consciousness is more likely to be helpful than a hierarchic approach.
Cross tabulations
We have provided some introduction to cross tabulations at reference 15. Such tabulations remain important in many branches of study, from the study of crime in Saxon Sussex to the study of jelly fish in the Java Sea, although they have moved from being the final product of much statistical work to being an intermediate product, with much of today’s final product coming from more or less sophisticated graphical presentation tools, rather than from the mundane cross tabulator. Which, to our minds, leads to a deplorable tendency to play with the graphical interfaces, rather than to look at the numbers.
 |
| Figure 7 |
There is a presumption that most properties are applicable to most members of the population, but note that it is easy enough to specify properties for which this is not true. Age of partner, for example, is a property which is only applicable to those members of the population with a partner. But it may, nevertheless, be an interesting property, perhaps in combination with age of subject. Location of tumour is another such, usually finessed by taking it into cause of death.
We then proceed along the lines set out in reference 16 and create an array, in which, for example, the first dimension is region of death and the second dimension is cause of death, this last grouped up in some convenient way, there being a lot more causes of death than can conveniently be presented on the page. An array with perhaps a hundred cells. We then count our millions of deaths into those hundred cells, often with the rule that each death goes into exactly on cell. The result is called a cross tabulation, a tabulation of one dimension or axis of analysis against another, with a small such tabulation shown in the figure above. The expectation is that this tabulation will reveal interesting relations between region and cause. It may turn out, for example, that cancer of the big toe is very common in the north, while inflammation of the liver is very common in the south – and this gives the doctor who want to reduce the incidence of both complaints something to work on.
 |
| Figure 8 |
 |
| Figure 9 |
A recap on the reference 4 collection of variables
Reference 4 came up with plenty of variables which one might use to analyse frames of consciousness, plenty of dimensions. An exercise which we believe has value, prompts us to ask interesting questions, even if it is not carried, as it was not there, into actual tabulations, cross or otherwise.
 |
| Figure 10 |
 |
| Figure 11 |
A way of describing dimensions which de-emphasises the numeric, linear view of a dimension, which, as already noted, are more appropriate for a variable like temperature than one like marital status.
 |
| Figure 12 |
These now twenty one binners are organised by the level (in LWS-N speak) at which they are bound and coloured red for classifying binners left, green for intermediate binners centre and blue for descriptive binners right. Note that the rather soft level, the take, attracts none of them. A softness identified in reference 7.
 |
| Figure 13 |
In any event a large collection of binners, and the ideal would be to collect enough data about enough frames of consciousness to be able to use any or all of them. One then does cross tabulations until you find the ones that were useful, that provide new insights into the organisation of consciousness.
Why are cross tabulations useful here?
Our particular Holy Grail is to come up with a satisfactory explanation of how consciousness comes to be, that is in the here and now, not in our developmental or evolutionary past, informative though those might be, and a bonus would be some explanation of what it is for. Such explanations needs to accommodate most of the many curious features and properties of consciousness and conscious behaviour that have been collected up over the years. We believe that suitable cross tabulations might be part of such an explanation.
The things to be cross tabulated are frames of consciousness, short periods of time when the subject holds his or her attention on a single object, with an object being anything that can hold attention, so an inclusive concept – with some indication of variety given in a list at reference 13. By short, we mean usually of the order of a second or so – but sometimes much longer, for example in the counting described at reference 8.
Reference 4, as we have indicated in the previous section, is full of stuff about how cross tabulations are or might be done, but does not so say so much about why they are useful – nor was the suggestion there of cross tabulating our descriptive experience sample pursued. But the core notion is that by looking at small, specialised groups of frames of consciousness, one can say something interesting about them. Rather as it is easier, certainly when starting out, to say something interesting about a rat or an elephant that it is to say something interesting about mammals in general. Mammals emerged as a top down construct which were fitted onto the results of bottom up analysis.
The Wikipedia article about what they call contingency tables at reference 17, points to some of the statistical tests which can be applied to test whether what look like connections between cells are really significant or whether the seeming connections could just as well be the product of chance, of random fluctuations in the data. In which we need to remember that if we have enough cells, then even random data is going to throw up some connections which look significant (for which problem ask Bing or Google for the Bonferroni correction ). In what follows we suppose that we are looking for fairly gross features of our data, for which statistical heavy-lifting is not necessary, at least for the present. A second line of defence is replication: if two independent bits of work on different datasets, different collections of frames of consciousness, come up with the same connection, then our confidence in that connection increases .
Classification by cross tabulation is also a vehicle for communication. One can talk about the classifications underlying each dimension of the tabulation and one can talk about the contents of the tabulation. One can speculate about which cells of the tabulation are going to be non-zero and which are going to be large. In an ideal world, one could compare cross tabulations drawn from different collections of frames of consciousness, most likely from different populations. Do classifications derived from experience of one world run in another?
Going further, one can talk about the neural machinery which underpins any particular cell in the tabulation, that is to say some group of frames of consciousness. The cross tabulation is one way of breaking the problem down into more manageable chunks. And while populating the cross tabulation helps, provides evidence about what is important, such population is not necessary for talk.
Reference 4 pointed to the fact that cross tabulations were more usually associated with more substantial, certainly less ephemeral and more accessible objects than frames of consciousness – things like munitions, households, people or cases of some complaint or disease.
In the case of people, one might be attempting to project the population into the future. Projections which depend on things like birth rates and death rates, both of which are far more stable if one can analyse one’s population by age, sex and various other demographic variables. So the birth rate of the population at large will vary a great deal over time, but the birth rate of married women between the ages of 20 and 24 will not.
In something of the same way, it may well be that there are interesting correlations between variables like age and sex and the incidence of particular kinds of frames of consciousness.
In the case of disease also, it is often useful to analyse the case (the disease) and the host (the person with the disease) together. Such analysis will often highlight the sort of people who are at risk and may, in consequence, help to identify underlying causes. In the past, for example, a great deal of effort has been put into analysing cause of death by occupation. Place of residence, used in our first simple example, has proved useful in identifying environmental or cultural factors – like the propensity to eat meat and potato pies or to smoke. Such cross tabulations are useful to the extent that the dimensions are not mutually independent, that there are cross correlations.
Note that while most people experience any one disease at most once, in the case of some particular kind of frame of consciousness, if they experience it at all, they may experience it many times. One might take an interest, for example, in the number of frames a day or correlations between particular kinds of frames of consciousness and particular kinds of day.
In analysing frames of consciousness, we trying to do at least three things.
Firstly, to build a classification which is applicable to frames of consciousness, in rather the same way as the more or less settled and stable classification applied to larger animals. A classification which is satisfying, neat and tidy. A classification which helps us talk about, to share information about the phenomenon of interest, in this case consciousness. The present proposition is that a multi-dimensional approach to classification is presently the best bet.
Secondly, we want the leaves of this classification, the terminal points of this classification to group together frames of consciousness which have a distinctive electrical signal which can be detected with an EEG machine or perhaps a brain scanner of one sort or another. Ducking for the moment the probability that some states of consciousness do not co-exist with present-day EEG machines, rather intrusive and cumbersome as they are.
Thirdly, it would be good if we could collect statistics, joining what we know about people’s outer properties and activities together with what we know about their inner lives. Perhaps by means of some extension of the Hurlburt project.
Some complications
 |
| Figure 14 |
We need here to mention another complication. In the example of a cross tabulation of deaths, we used single valued properties these deaths to define our two dimensions, our two axes. There is, for example, just one value of age at death. Each death is counted into exactly one place in the cross tabulations, totals and subtotals aside.
However, some statisticians find it useful to have properties which are not single valued. So, for example, medical statisticians often work with multiple causes of death. So somebody might die of pneumonia, but there was also an underlying cancer which had weakened the body, compounded by cirrhosis of the liver. Such a somebody might be given three causes of death.
Another way that multi-valued properties might arise is when the subject of a cross tabulation has child records. With the obvious example being the birth children of a mother: we are looking at the mother, but we want to bring information about her children into the analysis. Number of children is single valued, but age of children is multi-valued. Diseases of children is likely to be many-valued.
Parent records are more straightforward. A child has just one mother, so drawing properties of the mother down into an analysis of the children does not present any problem in this regard.
All this by way of a reminder that we need to keep in mind which population it is that we are tabulating.
Coding complications
What follows is a small sample of the sort of complications which can arise in coding. Classification variables are not always as straightforward as at first might appear.
 |
| Figure 15 |
Location in skull. We are collecting information about voxels in the heads of a number of subjects. We are interested in the location of those voxels. With what frame of reference do we locate those voxels? Some frame of reference derived from a scanner? Euclidean coordinates with respect to some standard orientation of the head? Euclidean coordinates with respect to some stretching of each particular head onto some standard head, possibly onto the plane, for the purposes of comparison? Coordinates which respect some standard identification of the many folds of the cerebral cortex – the part of the head where most of the voxels of interest lie? Much serious work has been devoted by brain scanning teams to these questions. While I remember that my father used to know a research dentist vexed by questions of this sort when taking photographs of the jaws of children at regular points through their childhood. .
Area of usual residence. We are collecting information about some population over some short period of time and one of the variables we are interested in is area. One of the members of our sample is a travelling salesman on his rounds. Do we count him where he was enumerated or where he usually lives? Or where he works out of? Another was a member of the armed forces on leave with his parents, about to go on deployment. Do we count him where he is based or where he is going to be for the next three months, possibly somewhere abroad, somewhere secret? All of which depends – at least to some extent – on whether we are an operations manager for McDonald’s wondering whether to open a new outlet in York or a care home operator wondering whether to build a new plant in Leeds.
Primary cause of death. Determining the best way to classify disease and death is an industry in its own right. And the result is a moving target, with WHO presently at the 11th revision of its international classification of diseases (ICD), with each revision is the result of a large number of international conferences of experts. And as regards mental disorders something called DSM-IV is maintained in the US – where classification is very important from the point of view of medical insurance. While forty years ago, in the UK, the big change was allowing death registration to include multiple causes of death.
It may well be that complications of this sort do not arise when tabulating frames of consciousness, but it is as well to bear them in mind. They are there and they might be nearer than one thinks.
Four dimensional graded consciousness
Both reference 1 (2017) and reference 3 (2012) present four dimensional analyses – analyses which are not quite the same. Things seem to have moved in in the intervening five years. There is additional material at reference 2 (2015).
The earlier paper, reference 3, opens with a distinction between just being conscious and being conscious of something. We suspect that this distinction is more a matter of how we choose to describe a frame of consciousness and we would argue that there has to be a something, there has to be an object of consciousness. Consciousness has to be about something, be it ever so woolly, be it some transcendent state of a Buddhist meditating on the void. And that something would be, in LWS-N, expressed as layer objects of one sort or another.
Reference 3 suggests that consciousness is indeed a unitary phenomenon, that all the different kinds of consciousness, all the different kinds of frames of consciousness, are instances of that one phenomenon, with the objective being to describe that phenomenon, to uncover its inner workings. To which end, those frames can usefully be analysed in a four dimensional framework.
 |
| Figure 16 |
Orders are ordered. We start with an organism being able to respond to a stimulus. Second, the organism is able to classify the stimulus. Perhaps to be able to say that this stimulus is a patch of the colour red. Third, the organism starts to organise the world in to objects. Perhaps to be able to say that this stimulus is a red post box (as they are in the UK). Fourth, we start to have consciousness of self. Fifth, we start to have consciousness of being conscious of something. And so it might go on, although in practise this seems unlikely, certainly marginal. Not the sort of thing that many people spend quality time on.
States is about physiological aspects of consciousness, some of which is captured by the three measures introduced above – Glasgow coma scale for one – and some of which is about impairments due to substances, damage or disorders.
Types is glossed in reference 1 along the lines: ‘Source: what is the major source of the information the subject is conscious of? Purpose: for what purposes, and in what circumstances, may the information given be made use of? Put another way, what is the aim and context-of-use for the consciousness? Subject: who or what is it that is conscious? Who or what is the subject? What type of animal or cognitive system can, and does, possess consciousness?’. All of which seems to cover a lot of ground. And not particularly ordered.
So the first dimension does not classify frames of consciousness at all. The second dimension addresses the issues around being conscious of being conscious, issues which philosophers like to talk about, and gives us an ordered classification with five bins. We are offered four bins for the third dimension, although we imagine it would not be difficult to come up with a finer grained classification. While we are offered what look to us like three more dimensions for the fourth dimension and no bins at all.
The later paper, reference 3, opens with what we regard as the sensible claim that a multi-dimensional framework might provide a good vehicle for describing and talking about states of consciousness. That reducing consciousness to a single number – say in the way of the Glasgow Coma Scale – is only going to capture one aspect of this complex phenomenon – one aspect which may well be important in its intended context, but which is far from the whole story.
This is followed by the claim that four is the proper number, that four dimensions float to the top of the vast literature which is accumulating on the subject.
 |
| Figure 17 |
 |
| Figure 18 |
The aspect abstractness/semantic seems to be very much the same sort of thing as the orders we had before. Say an integer in the range [1, 5]
From orders I associate to Gödel and the way he was able to describe the rules of arithmetic, of arithmetical proofs using the very same arithmetic. He was, for example, able to associate a proof with a positive integer. Which is where I hope to get to with LWS-N: we just have one data structure built around layers, layer objects and their parts, which can cope with all these orders without needing to get into philosophical niceties or special data.
The aspect complexity/physiological seems to be about the machinery needed to produce the state in question. Also about the neural correlates of the state in question, not quite the same thing. Nothing to with things like the Glasgow Coma Scale already mentioned, a scale which one might have thought had a strong claim to dimensionality – but which is specifically excluded in the discussion section. On the other hand, the PCI mentioned above is included, as is Tononi’s Φ. Both the PCI and Φ might be construed as properties of a frame of consciousness, properties one might address in a cross tabulation.
One can even see how one might measure PCI at the point that one asks the subject to report on his current frame of consciousness – although one might worry whether these two activities – measuring and reporting – interacted in an unhelpful way.
But quite different from the third dimension of the earlier paper. Perhaps a real in the range [0, 1].
While the aspect usefulness/functional seems to be very much the same sort of thing as the types we had before.
One of the examples the paper uses to illustrate the application of these dimensions is the phenomenon of being locked in: the first thee dimensions all score high, while the fourth scores low. Another is blindsight, where the first three dimensions all score low, while there is some action on the fourth dimension. With both examples suggesting that all four dimensions are to be regarded as both graded and ordered. While what we see is two dimensions of the [0,1] real interval variety, one of the {1, 2, 3 …} small integer sequence variety and one which does not seem to be very easy to order at all.
The impression given is that the authors are interested in dimensions and the gross structure of consciousness, but are not, at least on the evidence of this paper, terribly interested in the contents of consciousness, in the sorts of things turned up by Hurlburt or in the sorts of things which might be found in a structure like LWS-N.
Conclusions
We find the analysis of consciousness into four dimensions interesting rather than conclusive.
The papers at references 1, 2 and 3 do not spend much space on the contents of consciousness, in LWS-N speak, in our layers, layer objects and parts of layer objects. We believe that there is room for work in this area.
They are also evidence of the way in which those working on consciousness do spend a lot of space on patrolling the boundaries, on seeing what happens at the edge of consciousness or when things start to go wrong, for one reason or another. These patrols have been very productive, but we believe, following Hurlburt, that there is room for more work on the normal.
Work which might, on both counts, take the form of cross tabular analyses of large numbers of frames of consciousness. Perhaps two dimensions in the first instance, as they look best on the page. And then, maybe, one would be in a better position to say which of the many dimensions suggested in one or other of references 1, 3 and 4 were the interesting or important ones. And from whose point of view.
References
Reference 1: Four-Dimensional Graded Consciousness – Jakub Jonkisz, Michał Wierzchoń and Marek Binder – 2017.
Reference 2: Consciousness: individuated information in action - Jonkisz, J. 2015.
Reference 3: Consciousness: a four-fold taxonomy – Jonkisz, J. – 2012.
Reference 4: http://psmv3.blogspot.com/2017/10/on-taxonomy-of-consciousness.html. A long post, a fair amount of which is devoted to the classification, in various ways, of frames of consciousness.
Reference 5: http://psmv3.blogspot.com/2018/03/descriptors.html. A post about the various emotions expressed and engendered by music.
Reference 6: http://psmv3.blogspot.com/2018/06/measuring-consciousness.html. A rather different sort of post, about what can presently be found out about the consciousness of someone by looking at them with electrodes from the outside.
Reference 7: https://psmv4.blogspot.com/2020/01/another-look-at-frames.html.
Reference 8: https://psmv4.blogspot.com/2020/01/on-as-aspect-of-attention-and.html.
Reference 9: https://www.glasgowcomascale.org/.
Reference 10: https://www.medtronic.com/covidien/en-us/products/brain-monitoring.html.
Reference 11: Using the perceptual awareness scale (PAS) – Kristian Sandberg, Morten Overgaard – 2015. The source for Figure 18 above.
Reference 12: A Theoretically Based Index of Consciousness Independent of Sensory Processing and Behavior - Adenauer G. Casali, Olivia Gosseries, Mario Rosanova, Mélanie Boly, Simone Sarasso, Karina R. Casali, Silvia Casarotto, Marie-Aurélie Bruno, Steven Laureys, Giulio Tononi, Marcello Massimini – 2013.
Reference 13: https://psmv3.blogspot.com/2018/08/of-cabbages-and-kings.html. See the section ‘Examples of things in universes’.
Reference 14: Synthesis of phylogeny and taxonomy into a comprehensive tree of life - Cody E. Hinchliff and others – 2015.
Reference 15: http://psmv3.blogspot.com/2017/01/progress-report-on-descriptive.html.
Reference 16: http://psmv4.blogspot.com/2020/02/a-sort-of-pivot-table.html.
Reference 17: https://en.wikipedia.org/wiki/Contingency_table.
Reference 18: http://psmv4.blogspot.com/2020/02/my-mobile-my-doctor.html.
Reference 19: http://psmv3.blogspot.co.uk/2018/01/an-introduction-to-lws-n.html. A little out of date, but hopefully it will serve here.
Reference 20: https://psmv4.blogspot.com/2019/06/a-further-update-on-seeing-red.html. Inter alia, a list of the other material which has been posted. Also a little out of date.
Group search key: sre.
No comments:
Post a Comment