Consciousness

All mimsy were the borough groves..

Consciousness emerges from TDE architectonics

This page explains how consciousness (the subjective mind, ie one's self) is an emergent* property of the TDE (Trilevel fractal architecture, linguistic operating system) - see figure 11.

Before launching into a formal analysis of TGT from the viewpoint of consciousness, it seems wise to ask some circumspect questions of a general nature. The most important attitude to have in research is to expect simplicity as per 'Occam's Razor'. For example, why should my introspective mind be, in essence, any more mysterious or complicated than an automated, self-triggering picture-taking device, eg a digitally controlled video cam(era re)corder? The principle of subjectivity, of being INSIDE the system box, and not outside it, is illustrated most simply by this kind of device, and its 'punk' grand-nephew, the ubiquitous camera phone. In other words, there are important elements of consciousness that are derivable from the mere physical act of being located inside an enclosure, and obtaining data from outside, via a windowing mechanism of some kind**.

According to the TDE paradigm, consciousness (TDE level 2) evolved for the benefit of animals, not humans. Humans are conscious because they are animals. The special gift from evolution to humanity is language, not consciousness.

"Anyone who has had a cat, dog or parrot for a companion would never talk such nonsense about animals not being conscious".

'Common-coding' principle
The cybernetic reformulation of William James' 'common-coding' problem is called PCT, or Perceptual Control Theory. As well as being the only robotics programming paradigm that works in a practical sense, it also happens to represent an exemplary solution to the ages-old mind-body (identity) problem. Powers' PCT was independently discovered by Uexkull (who called it umwelt) in the 1930's, and Dyer (who called it Situation Image) in 2011.

Even after understanding subjectivity, and resolving the identity problem, one is still not 'out of the woods'. The next step is to gain insight into the dimensionality of subjective space, namely, that each agent's inner realm consists of two orthogonal (independent) measures, awareness/consciousness and agency/volition. This part of the overall picture arises from attempts to resolve Libet's Paradox. Part of the problem is that we use the word 'consciousness' to refer to two different but related things -(a) the whole subjective space, and to (b) one of the two dimensions that are needed to properly characterize that space.

The single greatest error made in thinking about consciousness is to ignore its efferent dimension.

*There is no intention to obfuscate - emergent properties are like the 'gestalt' of a composite image, they require all of the constituents at that level to be present, the image cannot be decomposed further without some loss of meaning.

**Barron, A.B. & Klein, C. (2016) What Insects can tell us about the origins of consciousness. PNAS vol. 113 : no 18 The essential point this article makes has been made before- that consciousness never occurs without those brainstem structures (or their morphological equivalents in insects) capable of representing the individual organism as a unified, subjectively controlled data structure of some kind.

Solution to Libet's Paradox

Libet's Paradox consists of a set of experimental results which show that the neural pre-motor processes for an individual's 'consciously willed' actions occur some 50 mSec or so BEFORE the person is consciously aware of making the decision to act. Until very recently, this finding seemed to provide evidence of a (clearly paradoxical) reversal of the flow of time - see Figure 32 below.

In Libet's Paradox, we see how making a bad implicit assumption in model construction can irretrievably muddy downstream waters, causing subsequent reasoning stages to collapse in paradox from which it is impossible to logically recover. The bad implicit assumption is this - the self, the agent in your head who makes your decisions, is identical to the conscious (accessible to introspection) part of your mind.

But how can this NOT be true? Any subconscious agent cannot be doing the bidding of the self, because the self is by definition the conscious 1/10 of the iceberg 'above the water line', if you will, and it is this conscious part (there is just one because there is only one self per person) which directs the subconscious parts as a master directs the slaves under his or her command.

Subjectivity = Volition + Consciousness
There is a rather surprising way out of this logical cul de sac which preserves the central and unique role of the self, and that is to realize that the subjective autonomous agency, or 'self' comprises not one but two such 'icebergs', quite separate but interconnected. The second aspect of self, the one that complements consciousness, is volition, or so-called 'free' will. Libet's Paradox is resolved without compromise by positing that subjectivity (reflexive and non-reflexive mental state-space of the self) can be both pre-conscious, yet voluntary. Agency is preserved yet experimental data is completely explained. In fact, according to this two-factor model of subjectivity, it is not the consciousness dimension but the volition dimension that most aptly represents the self's powers of agency, such as decision-making capacity, and belief-state self-attribution.

As we have seen with Libet's Paradox, reasoning about subjectivity (a.k.a phenomenal mind states, quality, psychophysics etc) in a manner consistent with the empirical data in this area is probably impossible without this higher-dimensional, two-factor orthogonalisation of the self-state-space.

This dramatic realisation, surely one of the most original acts of discovery in recent cognitive science history, leads to further, equally profitable insights. If, as is reasonable, consciousness is viewed as a superliminal state of self-awareness (capacity to perceive, judge and react) then volition can likewise be viewed as a superliminal state of self-agency (capacity to think, decide and behave). It follows that we should perhaps model ALL mental events in this manner, by separating those events that belong to the 'consciousness iceberg' (ie the thresholded INPUT channel) from those events that belong to the 'volition iceberg' (ie the thresholded OUTPUT channel). The end result of this reasoning process is the subjective view of cognitive computing, depicted in figure 5(a).

The smart way to view this revelation is by fitting it into the Universal Governance Paradigm (UGP). The pre-conscious period where the conscious intention is building up corresponds to the feedforward 'loading' phase of the UGP, while the subject's conscious experience of attending to the rotating timer in Libet's experiment corresponds to the 'interacting' phase, ie the feedback loop.

In fact, almost any effector (neural activation) process one wishes to choose seems to fit into this paradigm. The inevitable conclusion one must come to is that the two-phase UGP is indeed a universal model of consciousness. Note that we have not only broken subjectivity up into volition and consciousness, but we have now established that they occur in two temporally separated sequences, with the loading phase preceding the interactive phase.

The mirror test reveals a very important feature of consciousness - it interpolates between posture points a.k.a. situation images. Lets recap the salient features of this amazing experiment in consciousness, one that anyone can do at home. Stand in front of a mirror, and look first at your left eye, then your right, then your left. You know your eyeballs are darting left and right, in that dance called the 'saccades' or saccadic motion - you can feel your EOMs (Extra-Ocular Muscles) working, so you KNOW your eyeballs are moving. But what do you notice in the mirror? No eyeball motion enters your visual consciousness at all. Here are two important clues, (a) that consciousness can be different for different sensory modalities, ie your vision yields no conscious motion (of your eyeballs) but your proprioception (of your eyeballs) tells you that motion IS occurring, and (b) therefore that, as Dennett keeps reminding us, consciousness is not this great untouchable spooky spiritual thing, but a series of programming tricks ('hacks') the brain performs for the purposes of processing information correctly, and like many other programming hacks, they can exhibit 'bugs' like the eyeball motion mismatch.

The way that the brain produces motion (of any of its devices or sub-systems) is

(a) by specifying position (ie effects) only - forces are produced as a cybernetic side-effect of the positional specification. In a way, the brain has inverted the usual role of force as cause and motion as its effect. The commanded position is produced by an initial feedforward (unconscious) step

(b) by not generating arbitrary positions, but only those which correspond to stable configurations, ie postures and/or SI's.

(c) Voluntary movements are produced by transitions between stasis points- repeatable intermediate postures (the word used by the computer animators is 'keyframes'). These keyframes correspond to gravitationally stable postures, or mnemonically reproducible situation-images (SI's). The SI is the TDE level 2 equivalent of the posture at TDE level 1. The voluntary positions (states, vertices in the state graph) are produced by a Moore machine, a synchronous, non-deterministic (c-machine*) type of finite state automaton, implemented as a ROM.

(d) Involuntary movements, similar to reflexes, act in such a way as to 'join the dots' between keyframe configurations- The involuntary positions (transitions, edges in the state graph) are produced by a Mealy machine, an asynchronous, deterministic (a-machine*) type of finite state automaton, usually implemented as a ROM.

(e) The brain produces consciousness as a side-effect of movement- by changing perceptual features (image properties) in a smooth and largely seamless manner between keyframes / posture points. Computer animators have also developed a name for this concept- 'in-betweening'.

* these automaton classes are attributed to Alan Turing, 'a' meaning 'automatic', 'c' meaning 'choice'

Consciousness has two faces- (i) waking/memory acquisition and (ii) sleep/memory consolidation.

Waking aspect of consciousness

Consciousness has been explained as in-betweening ('filling-in' or interpolating) between stable 'keyframe' posture points/ situation images. The purpose of this waking aspect of consciousness is to (a) supply input symbols to the non-deterministic state machine responsible for voluntary action, causing it to become temporarily deterministic, and to proceed automatically until it reaches the next 'choice' input point. The synchronous Moore machine requires top-down choice in order to work, hence Turing's labelling it as the 'c-machine' (choice machine). That is, it is the user's goal-oriented decision that provides the input symbol which selects which of several states occur at the next 'clock' pulse (or equivalent biological trigger).

(b) fill in the gaps smoothly, where the subject's brain creates a seamless surface perception of reality, while transitions in the state machine occur in the background. The classic example of this is the 'phi' illusion, as illustrated in the following video:-

The mechanism behind the 'phi' illusion is the interpolation, or 'inbetweening' of the keyframes used by the brain to create voluntary motion. A more complex example is shown in the second video on this page. The ubiquitous subject S has been told to look in a mirror, while focussing both his eyes on the mirror images of his left and right eyeballs repeatedly in an alternated manner, L-R-L-R- etc. The video (titled 'phi-balls illusion') clearly shows this side-to-side motion, while S looks at his left eye, then his right eye (that is, at the images of his eyes in the mirror) repeatedly*.

*NOTE - this side-to-side eyeball flicking is a deliberate action, IT IS NOT SACCADIC, however, the same underlying mechanism explains why we are (almost) never aware of the constant saccadic motions of the eyeballs.

Subject sees no eyeball movement at all, even though he knows he is moving them

When the subject followed the instructions, what do you think he saw? NO EYEBALL MOTION AT ALL ! The explanation for this apparent conundrum (it really is amazing- you MUST try it yourself) is just as unexpected- we see what we EXPECT to see. The subject's eyeballs form part of the self-image that is formed by reflection in the mirror. Apart from the moving eyeballs, the reflected image is stationary, since the subject has not moved during the test. The brain EXPECTS no spatial movement of the image, so that is precisely what it sees. Hence no eyeball motion is detected visually, although the subject can feel the eyeball motion proprioceptively. The subject has then checked it objectively by making this short video with a smart phone*. If you think these effects are rather specialized, and don't apply to everyday life, you are seriously mistaken. All of us see what we EXPECT to see, no more, no less, as the following video shows :-

*what is so surprising is that such a 'serious' experiment on a topic as profound as consciousness can be performed using the most basic equipment by anyone at home.